¾ú¦~¬ã¨s­pµe·§ªp

±i´¼¬P


²ªí

¤¤¤å¦WºÙ ­^¤å¦WºÙ ¥D«ù¤H ¸É§U³æ¦ì ­pµe½s¸¹ ¸É§Uª÷ÃB ¶}©l¤é´Á µ²§ô¤é´Á
¥Í¦¨¦¡¹ï¸Ü¤§»OÆW¤H¤åª¾Ãѱ´°É¨t²Î­pµe  TAIHUCAIS: TAIwan HUmanities Conversational AI Knowledge Discovery System  ±i´¼¬P  °ê¬ì·|    2000000  2024/1/1  2026/12/31 
¥H¾÷¾¹¾Ç²ß¤èªk«Øºc°]°È¹w´ú¼Ò«¬  Revenue forecasting based on machine learning techniques  ±i´¼¬P  Ápµo¬ì§Þ    1200000  2023/11/1  2024/10/31 
¨Ï¥Î¨ã¦³¦h¼ÒºA¥Íª«ÃѧO¯S¼xªº¾÷¾¹¾Ç²ß¼Ò«¬¨Ó¹w´ú¤Úª÷´Ë¯fªº­·ÀI©M¯fµ{    ±i´¼¬P  °ê¬ì·|  112-2221-E-002-188-MY3  928000  2023/8/1  2026/7/31 
¶°µ²´¼¼z¦ç¤Î¼Æ¾ÚÀ³¥Î©ó¹B°Ê´_°·»Pºë·Ç¹B°Ê¬ãµo­pµe  The Integration of Smart Garments and Data Applications in the fields of Sports Rehabilitation and Precision Exercise  ±i´¼¬P  °ê¬ì·|  NSTC 112-2622-E-002-009  1376504  2023/6/1  2024/5/31 
¶}µoÁ{§É§Y®ÉÀË´ú»P¹w«á¨t²Î©óºC©ÊµÇŦ¯fªººë·Ç¶EÂ_»PªvÀø(2/3)    ±i´¼¬P  °ê¬ì·|  112-2218-E-002-055-  1450000  2022/8/1  2025/7/31 
AIÁn¾Ç½Õ­µ»Pµû´ú¨t²Î  AI Acoustic Evaluation System  ±i´¼¬P  µØºÓ¹q¸£    1300000  2021/11/1  2022/10/31 
»y­µ¿ëÃѼҲն}µo  Speech Recognition for Chinese News from Mainland  ±i´¼¬P  °ê®a°ª³t¹q¸£¤¤¤ß    1290000  2021/10/1  2022/9/30 
¾A¥Î©ó´¼¼zªA°Èªº¥i«H¿àAI¥ý¶i§Þ³N¬ã¨s¡G¡]¤l­pµe¤@¡^¨ã¥i«H«×»P¥i¾a«×ªº¦¸¥@¥NAI«ÈªA  Advanced Technologies for Designing Trustable AI Services  ±i´¼¬P  °ê¬ì·|    1500000  2021/10/1  2025/9/30 
ºqÁn¤ÀÂ÷ DNN model À³¥Î©ó TV SOC ¥­¥x  Singing Voice Separation via DNN Models on TV SOC Platform  ±i´¼¬P  ·ç¬R¬ì§Þ    1500000  2021/6/1  2021/5/31 
¦Û°Ê¤Æ­Ó¤H¹q¸£«~½è±±ºÞ´ú¸Õ¨t²Î  Automatic PC quality control and testing systems  ±i´¼¬P  ©MºÓÁp¦X¬ì§Þ    1640347  2020/11/1  2021/10/31 
«Ø¥ß¤H¤u´¼¼z¦h­±¦V¤ÀªR¥­¥x¥H¹w´ú¡B¶EÂ_¤Î¤¶¤J¤Úª÷´Ë¤ó¯g»P¬ÛÃö¯«¸g°h¤Æ©Ê¯e¯f  Developing an artificial intelligence-based multidomain sensing platform to predict and early intervene Parkinson's disease in aging society  ±i´¼¬P  ¬ì§Þ³¡  MOST 109-2221-E-002 -163 -MY3  1004000  2020/8/1  2021/7/31 
¾A¥Î©óÃä½t¹Bºâªº³ê¿ôµü¿ëÃѨt²Î  Keyword spotting on edge devices  ±i´¼¬P  Ápµú¬ì§Þ    1288000  2020/6/1  2021/5/31 
´Ó°ò©ó¹q¸£µøı¤§µL¤H¾÷Á×»Ù¬ã¨s­pµe  UAV obstacle avoidance system with computer vision  ±i´¼¬P  ¤¤¬ì°|    900000  2020/1/1  2020/12/31 
AI Cup (ºqÁnÂàÃлP©M©¶¿ëÃÑÄvÁÉ)  AI Cup: Singing Transcription and Music Chord Recognition  ±i´¼¬P  ±Ð¨|³¡    1600000  2019/11/1  2021/4/30 
¯¼Â´·~¤u·~4.0AI´¼¼z¶³ºÝ¦â®wº[¬V¦â°t¤è´¼¼zÀ³¥Î  Textile Industry 4.0: AI Intelligent Color Library on Cloud and Intelligent Application of Dyeing Formulas  ±i´¼¬P  ¬ì§Þ³¡  MOST 108-2745-8-002-007  1500000  2019/11/1  2020/10/31 
¤H¤u´¼¼z©óª÷¿Ä¬ì§Þ¤§À³¥Î  Applications of AI over FinTech  ±i´¼¬P  ¥É¤s»È¦æ    10000000  2019/6/1  2021/5/31 
S3: ¥þ¤è¦ì­µ¼Ö®T¼Ö¾Ç²ß¥­¥x (2/2)  S3: A Universal Music Platform  ±i´¼¬P  ¬ì§Þ³¡  MOST 107-2823-8-002-009  12000000  2018/12/1  2019/8/30 
´Óª«¼v¹³¿ëÃÑ  Plant Image Recognition  ±i´¼¬P  ¤u¬ã°|    1000000  2018/1/1  2018/12/12 
µ²¦X¤H¤u´¼¼z»Pª«Ápºô¬ì§Þµo®iºë·ÇºÎ¯vÂå¾Ç¡]¤À¶µ¤@¡Gµo®i©~®aºÎ¯v©I§l¤¤¤î¿zÀ˼ҫ¬¡^  Developing Artificial Intelligence and IOT Technology on Precision Sleep Medicine  ±i´¼¬P  ¬ì§Þ³¡    2000000  2018/1/1  2018/12/31 
»O¤jÂ寫¡Vºë·ÇÂåÀø¤H¤u´¼¼z»²§U¨Mµ¦¨t²Î¡m¤l­pµe¥|¡Gºë·ÇÂåÀø¤H¤u´¼¼z¶}µo¡^  NTU Medical Genie ¡V AI Decision Support System for Precision Medicine (Subproject 4: AI Technologies for Precision Medicine)  ±i´¼¬P  ¬ì§Þ³¡  MOST 107-2634-F-002-015  3800000  2018/1/1  2018/12/31 
S3: ¥þ¤è¦ì­µ¼Ö®T¼Ö¾Ç²ß¥­¥x (1/2)  S3: A Universal Music Platform  ±i´¼¬P  ¬ì§Þ³¡  MOST 106-3114-8-002-003  15000000  2017/12/1  2018/11/30 
À³¥Î²`«×¾Ç²ß¤èªk¤§»yªÌÅçÃÒ§Þ³N  Speaker Verification Using Deep Learning  ±i´¼¬P  ¤¤µØ¹q«H    990000  2017/10/20  2018/10/19 
´Óª«¼v¹³¿ëÃÑ  Plant Image Recognition  ±i´¼¬P  ¤u¬ã°|    990000  2017/9/1  2018/8/31 
¨t²Î©Ê¦Û°Ê¤Æ¶K¼Ð  Systematic Automatic Labeling  ±i´¼¬P  ¥É¤s»È¦æ    630000  2017/5/1  2017/10/31 
ºë·Ç¦æ¾P»P¨t²Î¤Æ¦Û°Ê¶K¼Ð  Precision Sales and Systematic & Automatic Labeling  ±i´¼¬P  ¥É¤s»È¦æ    600000  2017/3/1  2017/10/31 
¬F©²¥¨¶q¸ê®Æ¤ÀªR¤u¨ã»P¥­¥x¤l¡G¤l­pµe¤@. ¥¨¶q»y­µ¸ê®Æ¤ÀªR  Big Speech Data Analytics  ±i´¼¬P  ¬ì§Þ³¡    4947298  2016/7/1  2017/6/30 
ºq¦±¦±­·¤ÀÃþ¾¹»P±¡¹Ò¤ÀÃþ¾¹¤§¬ã¨s  Research on Music Genre Classification and Scenario classification  ±i´¼¬P  ¤u¬ã°|    500000  2016/4/1  2016/11/15 
¥Î©ó­µ¼Ö¸ê°TÀ˯Áªº²`«×¾Ç²ß  Deep Learning for Music Information Retrieval  ±i´¼¬P  ¬ì§Þ³¡    2938000  2015/8/1  2018/7/31 
­µ°T«ü¯¾À˯Á»P¯B¤ô¦L´O¤J§Þ³N  Audio Fingerprinting & Audio Watermarking  ±i´¼¬P  ¤¤µØ¹q«H    980000  2015/4/1  2016/3/31 
§Ü¤zÂZªº­µ°T¸ê°TÁôÂç޳N  Noise-robust Information Hiding for Audio Signals  ±i´¼¬P  ¸êµ¦·|    600000  2015/1/1  2015/12/15 
Layout Sensitivity Model for NTO/CDU APC  Layout Sensitivity Model for NTO/CDU APC  ±i´¼¬P  ¥x¿n¹q    900000  2014/5/15  2015/5/14 
¤j¶q´CÅé¯S¼x¸ê®Æ®w¤ñ¹ï·j´M§Þ³N  Matching and Retrieval Techniques for Large Multimedia Databases  ±i´¼¬P  ¸êµ¦·|    800000  2014/2/1  2014/12/15 
´¹¶ê¯Ê³´¼Ë¦¡¿ë»{¤Î¬Û¦ü«×¤ÀªR  ¢åafer Failure Pattern Fingerprint and Similarity Detection  ±i´¼¬P  ¥x¿n¹q    500000  2013/9/1  2014/8/31 
¥HGPU¬°¹Bºâ®Ö¤ß¤§­µ¼ÖÀ˯Á¨t²Î  GPU-based Music Information Retrieval Systems  ±i´¼¬P  °ê¬ì·|    795000  2013/8/1  2015/7/31 
¦Û°Ê¤º®e¿ëÃѧ޳N  Automatic Content Recognition Technology  ±i´¼¬P  ¸êµ¦·|    800000  2013/2/1  2013/12/15 
±q¼Æ¦ì¾Ç²ß¨ì´¼¼z¥Í¬¡ªº¾ã¦X¬ãµo­pµe  Integrated R&D Program from Digital Learning to Smart Life  ±i´¼¬P  °ê¬ì·|    1000000  2013/1/1  2015/11/30 
´¹¶ê¯Ê³´¼Ë¦¡¿ë»{  Wafer Failure Pattern Fingerprint  ±i´¼¬P  ¥x¿n¹q    740000  2013/1/1  2013/12/31 
¤f»¡¥x»yµû¤À¨t²Î¤§¬ã¨s»P¹ê§@  Research and Implementation of Spoken Taiwanese Scoring System  ±i´¼¬P  °ê¬ì·|    523000  2012/8/1  2013/7/31 
±m§©Âಾªº¹ê§@»P±´°Q  The Practice and Implementation of Makeup Transfer  ±i´¼¬P  °ê¬ì·|¡B³Ð·NÅÚ½³    900000  2012/6/1  2013/5/31 
­µ¼Ö¶i¶¥¯S¼x©â¨ú»P¤H¾÷¤¬°Ê§Þ³N  Technologies for Advanced Music Feature Extraction and Human¡VComputer Interaction  ±i´¼¬P  ¤¤µØ¹q«H¬ã¨s©Ò    980000  2012/1/1  2012/12/31 
³z¹L»y­µ»PÃöÁä¦r²Õªº¹qµø¸`¥Ø¦Û°Ê¸ê®ÆµÑ¨ú¤èªk  Method for Automatic Data Extraction of TV Programs by Voice and Keyword Groups  ±i´¼¬P  ¸êµ¦·|    600000  2012/1/1  2012/12/15 
´¹¶ê¯Ê³´¹Ï¹³¤§¤ÀªR»P¿ëÃÑ  Wafer Failure Pattern Recognition  ±i´¼¬P  ¥x¿n¹q    600000  2011/9/1  2012/8/31 
¤À´²¦¡¶³ºÝ¤¤¤¶³nÅé¡]E2¤l­pµe¡G¶³ºÝÀ³¥Î§G¸p»PºÞ²z¨t²Î¡^  Cloud Application Deployment and Management System  ±i´¼¬P  ¥x¹F¹q    700000  2011/8/1  2014/7/31 
°ò©ó¼v¹³¤§Áy³¡½§½è¤ÀªR»P·å²«­×¸É  Image-based facial skin analysis and flaws covering  ±i´¼¬P  °ê¬ì·|¡B³Ð·NÅÚ½³  NSC 100-2622-E-007 -007 -CC2  700000  2011/6/1  2012/5/31 
¤ä´©Cloud-aware´O¤J¦¡¦æ°Ê¦h®Ö¤ß¥­¥x--¤l­pµe¤T¡G¾ã¦X´O¤J¦¡¨t²Î»P¶³ºÝ­pºâªº­µ¼Ö»P»y­µªA°È  Supporting voice/music services for mobile & cloud synergism  ±i´¼¬P  °ê¬ì·|  NSC 100-2219-E-007 -008  800000  2011/5/1  2013/4/30 
®É¶¡§Ç¦C¦æ¬°±´°É§Þ³N  Temporal and Sequential Activity Mining  ±i´¼¬P  ¸êµ¦·|    600000  2011/1/1  2011/12/31 
¥x»y»y­µ»P¤å¦r¦h­±¦V»y®Æ®w¤§«Ø¸m¤Î¨ä¦b¥x»y­pºâ»y¨¥¾Ç¤§À³¥Î--¤f»¡¥x»yµû¤À¨t²Î¤§¬ã¨s»P¹ê§@  Corpus Collection for Taiwanese Texts and Speech with Applications to Taiwanese Computational Linguistics - The Research and Development of Spoken Taiwanese Scoring Systems  ±i´¼¬P  °ê¬ì·|  99-2221-E-007-049-MY3  600000  2010/8/1  2013/7/31 
°ò©ó¼Ò¦¡ÃѧO¤èªk¶i¦æ¹q¾¹¯Ó¯à¯S¼x¤ÀªR  On the Use of Pattern Recognition Methods for Household Appliance Modeling Based on Readings of Electricity Meters  ±i´¼¬P  ¸êµ¦·|    600000  2010/3/1  2010/12/31 
­ó°Û·j´M§Þ³N  Techniques for Query by Singing/Humming  ±i´¼¬P  ¸êµ¦·|    660000  2010/2/1  2010/12/31 
¾A¥Î©ó´O¤J¦¡¨t²Îªº¹q¸£»²§U¤f»¡µØ»yµo­µ½m²ß¨t²Î  Computer Aided Spoken Chinese Pronunciation Practice System for Embedded Systems  ±i´¼¬P  °ê¬ì·| & Üg«ä¬ì§Þ    403000  2009/11/1  2010/10/31 
¥Î©ó­µ°T­µ¼Öªº­pºâťı³õ´º¤ÀªR¡]3/3¡^  Computational Auditory Scene Analysis for Audio Music  ±i´¼¬P  °ê¬ì·|  NSC 96-2628-E-007 -141 -MY3  521000  2009/8/1  2010/7/31 
¥xÆW¦Û¥D³B²z¾¹Android¥­¥x²`¯Ñ­pµe  Deep Cultivation for Taiwan's Independent Processor for Android Platforms  §õ¬F±X  ¸gÀÙ³¡¾Ç¬ã­pµe    800000  2009/6/1  2010/5/31 
°Û§@­Ñ¨Î¦³Án®Ñ¹q¤l¤½¥J­pµe  Excellent Singing and Reading Electronic Figures for Audiobooks  ª÷¥ò¹F  ¸gÀÙ³¡¾Ç¬ã­pµe    800000  2009/6/1  2010/5/31 
¥H»yªÌ¿ëÃѬ°°ò¦¤§´¼¼z«¬¤H¾÷¤¶­±  Intelligent Man-machine Interface based on Speaker Recognition  ±i´¼¬P  ¸êµ¦·|    600000  2009/3/1  2009/12/31 
±q»y­µ¹ï¸Ü¶i¦æ±¡ºü¿ëÃÑ  Emotion Detection from Spoken Dialog  ±i´¼¬P  ¸êµ¦·|    800000  2009/3/1  2009/12/31 
´O¤J¦¡¦h®Ö¤ß½sĶ¾¹»PÀ³¥Î³nÅ饭¥x¬ãµo­pµe  R & D for Embedded Multi-core Compiler and Application Software Platform  §õ¬F±X  ²MµØ¤j¾Ç    1940000  2009/3/1  2010/12/31 
´O¤J¦¡²§¦h®Ö¤ß¨t²Î§Þ³N¬ãµo3¦~­pµe(²Ä2´Á)  Embedded heterogeneous multi-core system technology research and development 3-year plan  ±i´¼¬P  ¸gÀÙ³¡¬ì±M­pµe    600000  2008/11/1  2010/10/31 
IntelÁp¦X¬ãµo­pµe  Intel Joint Research and Development Program  ±i´¼¬P  Intel    600000  2008/8/1  2009/7/31 
¥Î©ó­µ°T­µ¼Öªº­pºâťı³õ´º¤ÀªR¡]2/3¡^  Computational Auditory Scene Analysis for Audio Music  ±i´¼¬P  °ê¬ì·|  NSC 96-2628-E-007 -141 -MY3  521000  2008/8/1  2009/7/31 
Ápµo¬ì´O¤J¦¡¨t²Î§Þ³N¬ã¨s¤Î¤H¤~°ö¨|­pµe¡]²Ä¥|¤l­pµe¡Gµø°T¤Î»y­µÀ³¥Î¶}µo¡^  MediaTek Embedded System Technology Research and Talent Cultivation Program (Fourth Sub-Program: Video and Voice Application Development)  ±i´¼¬P  Ápµo¬ì    600000  2008/8/1  2009/7/31 
Tri-toneªº³sÄòÁn½Õ¶ì¼Ò¤Î°»¿ù§Þ³N  Tri-tone Based Continuous Tone Modeling and Analysis  ±i´¼¬P  ¸êµ¦·|    600000  2008/3/1  2008/11/30 
»y­µ¿ëÃѨt²Î¶}µo  Speech Recognition System Development  ±i´¼¬P  ¤¤¬ì°|    800000  2008/2/1  2008/11/30 
¥Î©ó­µ°T­µ¼Öªº­pºâťı³õ´º¤ÀªR¡]1/3¡^  Computational Auditory Scene Analysis for Audio Music  ±i´¼¬P  °ê¬ì·|  NSC 96-2628-E-007 -141 -MY3  521000  2007/8/1  2008/7/31 
µØ»y¤å¤¬°Ê»y­µ±Ð¾Ç§Þ³N¬ãµo  Speech-based Dialog Technologies for Learning Mandarin Chinese  ±i´¼¬P  ¸êµ¦·|    500000  2007/5/1  2007/12/31 
¥H»y­µ¿ëÃѤΦX¦¨¬°°ò¦ªº¤f»¡µØ»y¤å¹q¸£»²§U¾Ç²ß¨t²Î  A Spoken Mandarin Learning System Based on Speech Recognition and Synthesis  ±i´¼¬P  °ê¬ì·|  95-3113-S-007-001-  1494000  2006/12/1  2007/11/30 
Å¥¼g¨t²Î¤§»yªÌ½Õ¾A§Þ³N  Speaker Adaptation for an Embedded Dictation System  ±i´¼¬P  ÁÉ·L¬ì§Þ    500000  2006/11/1  2007/10/31 
­µ¼Ö·j´Mªº¥[³t»P¿ëÃѲv´£¤É¡A¤Î¨ä¦b´O¤J¦¡¨t²Îªº¹ê§@»PÀ³¥Î¡]3/3¡^  Speedup Mechansim and Performance Enhancement for Music Information Retrieval, with Applications to Embedded Systems  ±i´¼¬P  °ê¬ì·|  95-2221-E-007-220-  662000  2006/8/1  2007/7/31 
PDA¤¤¤å»y­µ¦X¦¨¨t²Î  Chinese TTS for PDA  ±i´¼¬P  ·L¬P¬ì§Þ    1000000  2006/7/1  2007/6/30 
±j°·©ÊÃöÁäµü»y­µ¿ëÃÑ  Robust Keyword Spotting  ±i´¼¬P  ·L¬P¬ì§Þ    1000000  2006/7/1  2007/6/30 
¤â«ù¦¡¸Ë¸mªº¤f»¡¤¤¤å»²§U¾Ç²ß§Þ³N  Computer-assisted Spoken Chinese Learning Systems for Hand-held Device  ±i´¼¬P  ¸êµ¦·|    500000  2006/3/1  2006/12/31 
¥xÆW¼Æ¦ì¦³Án®Ñºô¤§«Ø¸m»P±À¼s  The Development and Promotion of the Web Portal for Digital Talking Books in Taiwan  ±i´¼¬P¡B­ð¶Ç¸q¡B³¯©yªY  °ê¬ì·|  NSC 95-2422-H-007 -001  1368140  2006/3/1  2007/2/28 
­µ¼Ö·j´Mªº¥[³t»P¿ëÃѲv´£¤É¡A¤Î¨ä¦b´O¤J¦¡¨t²Îªº¹ê§@»PÀ³¥Î¡]2/3¡^  Speedup Mechansim and Performance Enhancement for Music Information Retrieval, with Applications to Embedded Systems  ±i´¼¬P  °ê¬ì·|  NSC 93-2213-E-007-058  695000  2005/8/1  2006/7/31 
§Q¥Î¥xÆW²{¦³µø»Ù¥Î¼Æ¦ì¨åÂøê®Æ»s§@ª¼¤H¹q¤l®Ñ-µo®iDAISY¤¤¤å¼½©ñ¾¹¤Î¬ÛÃö¤§»y­µ¿ëÃÑ»P¦X¦¨¥\¯à(¤l­pµe¤G)  On the Development of DAISY Chinese Player with Speech-enabled Interface Based on Speech Recognition and Synthesis  ±i´¼¬P  °ê¬ì·|  NSC 94-2422-H-007-005  919600  2005/3/1  2006/2/28 
¦h¼Ò¦¡­µ¼ÖÀ˯Áªº¥[³t¤èªk  Speedup Mechanisms for Multi-modal Music Information Retrieval  ±i´¼¬P  °ê¬ì·|¡B²M½«¬ì§Þ  NSC 93-2622-E-007-012-CC3  327000  2004/11/1  2005/10/31 
¥xÆWµø»Ù¥Î¼Æ¦ì¨åÂ䧻y­µÀ˯Á¨t²Î  Speech-based Information Retrieval for the Blind by Digital Archives in Taiwan  ­ð¶Ç¸q¡B±i´¼¬P  °ê¬ì·|    1000000  2004/8/1  2005/4/30 
­µ¼Ö·j´Mªº¥[³t»P¿ëÃѲv´£¤É¡A¤Î¨ä¦b´O¤J¦¡¨t²Îªº¹ê§@»PÀ³¥Î¡]1/3¡^  Speedup Mechansim and Performance Enhancement for Music Information Retrieval, with Applications to Embedded Systems  ±i´¼¬P  °ê¬ì·|  NSC 93-2213-E-007-058  695000  2004/8/1  2005/7/31 
À³¥Î©ó¼Æ¦ì±Ð§÷ªº»y­µÀ˯Á»P»yªÌ½T»{  On the Use of Speech-based Retrieval and Speaker Verification for Digital Courseware  ±i´¼¬P  ²Î«H¥ø·~ºÞ²zÅU°Ý    500000  2004/3/1  2004/10/31 
¤j«¬­µ¼ÖÀ˯Á¨t²Îªº²z½×»P¹ê§@¡]3/3¡^  Large-scale Music Information Retrieval System: Theory and Implementation¡]3/3¡^  ±i´¼¬P  °ê¬ì·|  NSC 90-2213-E-007-058  882300  2003/8/1  2004/7/31 
­µ¼ÖÀ˯Áªº¥[³t¤èªk  Methods for Efficient Music Retrieval  ±i´¼¬P  °ê¬ì·|¡B²M½«¬ì§Þ    500000  2003/6/1  2004/5/31 
­µ°T§Þ³N¬ã¨s  Studies on Audio Technology  ¾G¤h±d¡B¿à­¸ò¼¡B±i´¼¬P¡BĬ¤åà±  ­â¶§¬ì§Þ    800000  2002/9/1  2003/8/31 
¤j«¬­µ¼ÖÀ˯Á¨t²Îªº²z½×»P¹ê§@¡]2/3¡^  Large-scale Music Information Retrieval System: Theory and Implementation¡]2/3¡^  ±i´¼¬P  °ê¬ì·|  NSC 90-2213-E-007-058  772300  2002/8/1  2003/7/31 
­µ¼ÖÀ˯Áªº¥[³t¤èªk  Methods for Efficient Music Retrieval  ±i´¼¬P  °ê¬ì·|¡B²M½«¬ì§Þ    500000  2002/6/1  2003/5/31 
»y­µ°T¸¹³B²z»P¿ëÃѪº³nµwÅé¹ê§@»P¾ã¦X  Audio Signal Processing and Recognition: Software/Hardware Implementation and Integration  ±i´¼¬P  Üg«ä¬ì§Þ    800000  2002/1/1  2002/12/31 
¤j«¬­µ¼ÖÀ˯Á¨t²Îªº²z½×»P¹ê§@¡]1/3¡^  Large-scale Music Information Retrieval System: Theory and Implementation¡]1/3¡^  ±i´¼¬P  °ê¬ì·|  NSC 90-2213-E-007-058  652300  2001/8/1  2002/7/31 
»y­µ»PºqÁn¦X¦¨  Speech and Singing Voice Synthesis  ±i´¼¬P¡B¶À²ÐµØ  ²M½«¬ì§Þ    600000  2001/8/1  2002/7/31 
»y­µ»P­µ¼Ö°T¸¹ªº¤ñ¹ï¤èªk»P¥[³t¾÷¨î  Music/Speech Information Retrieval and Their Speedup Mechanisms  ±i´¼¬P  ²M½«¬ì§Þ    600000  2001/5/1  2002/7/31 
¥H¤º®e¬°¥Dªº¦h´CÅéÀ˯Á¨t²Î ¢w ¥Ñ³nÅé¨ìµwÅ骺§Ö³tÂú§Î»Pµo®iÀô¹Ò  A Fast Prototyping Environment for Content-based Multimedia Information Retrieval  ±i´¼¬P  Üg«ä¬ì§Þ    800000  2000/12/1  2001/12/31 
­µ¼ÖÀ˯Á§Þ³Nªº¥[³t»P§ï¨}  Efficient and Effective Techniques in Music Information Retrieval  ±i´¼¬P  ²M½«¬ì§Þ      2000/9/1  2001/8/31 
²{¦³¼v¹³/»y­µ³B²z¬ã¨s¨å½d©óÂåÀø¦Û°Ê¤Æ¤§¯Ê¥¢»P§ï¶i - ¥H®ÖºÏ¦@®¶ÂåÀø»²§U¶EÂ_¨t²Î¤§¼v¹³³B²z§Y®É¤Æ¤Î¿é¤J»y­µ¤Æ¬°®×¨Ò  On-line Image Processing and Voice Activation in Magnetic Resonance Computer-Aided Diagnosis: A Strategy for Overcoming the Limitations in State of the Art Signal Processing Techniques  ³Å®a±Ò¡B´^®¶¿³¡B±i´¼¬P  °ê¬ì·|      2000/8/1  2001/7/31 
¾ã¦X»y­µ¿ëÃÑ»P¦X¦¨ªº¥Hºq¿ïºq¨t²Î  On the Integration of Speech Recognition/Synthesis into a Content-based Music Retrieval System  ±i´¼¬P¡B¶À²ÐµØ  ²M½«¬ì§Þ      2000/8/1  2001/7/31 
²MµØ¤j¾Ç¡u«D¦P¨B¤Þ¾É¦¡»·¶Z±Ð¾Ç¨t²Î¡v±À°Ê­pµe  Asynchronous Distance Learning Based on a Web Guiding System  ¤ý¤p¤t¡B¶À¤@¹A¡B±i´¼¬P¡B·¨¨û­ë  ±Ð¨|³¡      2000/1/1  2000/12/31 
¥HÁn¯¾»P¤HÁy¬°¥Dªº¥Íª«»{µý¨t²Î  Biometric Identification System Based on Face and Voice  ±i´¼¬P  ¥îÂ׬ì§Þ      1999/10/1  2000/7/31 
ª½¬yÅÜÀWªÅ½Õ¾÷±±¨îµ¦²¤µo®i  Design and developments of the control strategy for a DC variable-frequency air conditioner  ±i´¼¬P  ¤u¬ã°|¯à¸ê©Ò      1999/10/1  2000/10/31 
¯«¸g¼Ò½k¨t²Îªº»~®t¹w¦ô»Pµ²ºc¿ëÃÑ  Error Estimation and Structure Identification of Neuro-fuzzy Systems  ±i´¼¬P  °ê¬ì·|  NSC 89-2213-E-007-067    1999/8/1  2000/7/31 
·s»D¤ÀÃþ»P¤å¥óºK­n§Þ³N¶}µo  Classification and Summarization for On-line News  ±i´¼¬P  ¤Ó¤@«H³q      1999/8/1  2000/7/31 
¸õÀWºô¥x¤ÀªR  The analysis of radio communication networks  ±i´¼¬P  ¤¤¬ì°|      1999/7/1  2001/6/30 
¦³Án¹q¤l¹Ï®ÑÀ]ªº¦Û°Ê¤Æ§Þ³N»P¤u¨ã  A Study of Techniques and Tools for Audio/Textual Digital Library  ±i«T²±¡B­ð¶Ç¸q¡B±i´¼¬P  °ê¬ì·|      1998/8/1  1999/7/1 
³n¦¡­pºâ¤¤ªº»~®t¹w¦ô»Pµ²ºc¿ëÃÑ  Error Estimation and Structure Identification in Soft Computing  ±i´¼¬P  °ê¬ì·|  NSC88-2213-E-007-007    1998/8/1  1999/7/1 
µL½u¹qºô¥x¤ÀªR  The Analysis of Radio Communication Networks  ±i´¼¬P  ¤¤¬ì°|      1998/8/1  1999/7/1 
ºô¯¸À˯Á·j´M¤ÞÀº»Pºô¸ô§Y®É·s»DªA°È  Web Search Engines and On-line News Service and Technology  ±i´¼¬P  ¤Ó¤@«H³q      1998/7/1   
ºô¸ô§Y®É·s»DªA°Èªº§Þ³N¶}µo  Web On-line News Service and Technology  ±i´¼¬P¡B±i«T²±  ¤Ó¤@«H³q      1998/3/1  1998/7/1 
²M½«¶éºô¸ô®Ñ°|µo®i­pµe  Cyber University  ¶À¤@¹A¡B·¨¨û­ë¡B±i´¼¬P  ±Ð¨|³¡      1998/1/1  1998/7/1 
¤À´²¦¡¦h¦øªA¾¹ÀH·Nµø°T¨t²Î(III)(¤l­p¹º¤T) ´¼¼z«¬¬d¸ß¨t²Î (¥Hºq¿ïºq)  An Intelligent Interface of Query by Singing in VOD (Video on Deman)  ±i´¼¬P  °ê¬ì·|  NSC87-2213-E-007-013    1997/8/1  1998/7/1 
³n¦¡­pºâ¦b¸ê®Æ¼Ò«¬¤ÆªºÀ³¥Î  Soft Computing in Data Modeling  ±i´¼¬P  °ê¬ì·|  NSC87-2213-E-007-009    1997/8/1  1998/7/1 
»yªÌ¿ë»{  Speaker Recognition  ±i´¼¬P  °ê¬ì·|  NSC 86-2213-E-007-048    1996/8/1  1997/7/31 

¸Ô²Ó¸ê®Æ

  1. ¥Í¦¨¦¡¹ï¸Ü¤§»OÆW¤H¤åª¾Ãѱ´°É¨t²Î­pµe

    • ­^¤å¦WºÙ: TAIHUCAIS: TAIwan HUmanities Conversational AI Knowledge Discovery System
    • ­pµe½s¸¹:
    • ¥D«ù¤H: ±i´¼¬P
    • ¸É§U³æ¦ì: °ê¬ì·|
    • ­pµe°õ¦æ´Á¶¡: 2024/1/1 to 2026/12/31
    • ÃöÁäµü: Machine learning, large language model, chatbot
    • ºK­n²¤¶:
      »OÆW¾Ö¦³¦UºØ¿W¦³¦Ó¬Ã¶Qªº¤H¤å¸ê®Æ¡AµM¤ÀÄݤ£¦P³æ¦ì¡A¨Ï¥ÎªÌ¥²¶·¶i¤J¤À¥ßªº¸ê®Æ®w¨ú¥Î¡C¦Óªñ¦~¨ÓÀHµÛ¥Í¦¨¹ï¸Ü¦¡¤j«¬»y¨¥¼Ò«¬ªº¥X²{¡Aª¾Ãѱ´°Éªº¤è¦¡¤]¥ÑÃöÁä¦r·j´M§ï¬°¦ÛµM»y¨¥´£°Ý»P¦h½ü¹ï¸Üªº¤è¦¡¨Ó¨ú±o¾ã¦Xªº¸ê°T¡CµM¦Ó¥Ø«e³q¥Îªº¤j«¬»y¨¥¼Ò«¬¥¼¯à²Å¦X¤H¤å±M·~ªº»Ý¨D¡A¥B¤¤¤åªº²£¥X¤]¥¼¥²¶Kªñ»OÆWªº»y¹Ò¡C¬°´£¤É»OÆW¤H¤å¸ê®ÆªºÀ˯Á¤è«K©Ê¥H¤Î¤H¤å¬ã¨sªÌ»P¤u§@ªÌªº±M·~»Ý¨D¡A¥»­pµeÀÀµ²¦X»OÆW²{¦³ªº²Ä¤@¤â¥þ¤å¤H¤å¸ê®Æ®w»P»OÆW¦Û«Øªº³q¥Î»y¨¥¼Ò«¬¡A«Ø¸m¤@­Ó¸ê®Æ®wÀ˯Á¿é¥X»P¹ï¸Ü¥­¥x¡C³o­Ó±M¬°»OÆW¤H¤å¾ÇªÌ¶q¨­­q°µªº¹ï¸Ü¦¡ª¾Ãѱ´°É¨t²Î¡]¥xÃ[¶}¥Ü¡^±N³sµ²²{¦³ªº¸ê®Æ®w¡B¥Í¦¨©Ò»Ýªº¸ê°T¡B¥H²Å¦X¥xÆWijÃD»P»y¹Òªº¹ï¸Ü¦^À³¡B¨ÃÀu¤Æ¨Ï¥ÎªÌªºª¾Ãѱ´°ÉÅéÅç¡C¨Ï¥ÎªÌ¤£¦A»ÝÁcº¾¦a¶i¤J¦U³æ¿Wªº¸ê®Æ®w¶i¦æÀ˯Á¡A¦Ó¯à¦b³æ¤@¤¶­±¤¤¥H¦ÛµM»y¨¥´£°Ý¨Ã©M¨t²Î¤¬°Ê¡A¨Ïª¾ÃѪº±´°É§ó¬°«K±¶¦³®Ä¡Bºë·Ç¡B¥B²Å¦X»Ý¨D¡C¹w´Á¦¹¨t²Î¯à´£°ª»OÆW¤H¤å¸ê®Æ®wªº¨Ï¥Î²v¡A¨Ã¯àÂǥѻP»y¨¥¼Ò«¬¹ï¸Ü¤¬°Êªº¯S©Ê¡A§l¤Þ§ó¦h¤HÃöª`¤F¸Ñ»OÆWªº¤H¤å¾Ç¦U­±¦V¡C¦¹¥~¡A±N¹ï¸Ü¦¡»y¨¥¼Ò«¬¥Ñ³q¥Î«¬Âର±M·~«¬¬O¥Ø«e¥@¬É¦U±M·~»â°ìªº­«­n¥Ø¼Ð¡A­Y¦¹±´°É¨t²Î«Ø¸m¦¨¥\±N·|¬O­«­nªº§Þ³N³Ð·s¡A¥i¨Ñ¨ä¥L±M·~»â°ì°Ñ¦Ò»PÀ³¥Î¡C
      Taiwan possesses various unique and invaluable humanities datasets; however, these are managed by different entities, requiring users to access separate datasets for the information they need. With the recent development of large language models (LLMs), users' knowledge discovery methods have shifted from keyword searches to natural language queries and multi-turn dialogues to obtain integrated information. However, current available general-purpose LLMs do not meet the specialized needs of humanities researchers, and their Chinese language output may not align with the Taiwanese context. To enhance the convenience of accessing Taiwanese humanities data and meet the demands of humanities researchers and professionals, this project intends to integrate Taiwan's existing first-hand full-text humanities datasets with a locally developed general-purpose language model in Taiwan. The aim is to establish a user-friendly data retrieval and dialogue platform for knowledge discovery. This TAIwan Humanities Conversational AI knowledge discovery System (TAIHUCAIS), tailored specifically for Taiwanese humanities data, will be better suited to generate information and engage in dialogues that align with the context of relevant Taiwanese topics. It will optimize user experience, eliminating the need for them to navigate individual databases separately. Instead, users will be able to pose questions in natural language through a single interface, with the system aggregating and presenting relevant information, making knowledge discovery more convenient, efficient, precise, and tailored to their needs. The anticipated outcome of this system includes increased use of Taiwanese humanities datasets and better engagement with more users in various aspects of Taiwan's humanities. Moreover, transitioning from a general-purpose language model to a specialized one is a vital objective across many fields worldwide. The development of TAIHUCAIS will not only contribute to significant technological innovations but also potentially establish a benchmark for other specialized domains, creating new opportunities for applications.

  2. ¥H¾÷¾¹¾Ç²ß¤èªk«Øºc°]°È¹w´ú¼Ò«¬

    • ­^¤å¦WºÙ: Revenue forecasting based on machine learning techniques
    • ­pµe½s¸¹:
    • ¥D«ù¤H: ±i´¼¬P
    • ¸É§U³æ¦ì: Ápµo¬ì§Þ
    • ­pµe°õ¦æ´Á¶¡: 2023/11/1 to 2024/10/31
    • ÃöÁäµü: Machine learning, revenue forcast
    • ºK­n²¤¶:

  3. ¨Ï¥Î¨ã¦³¦h¼ÒºA¥Íª«ÃѧO¯S¼xªº¾÷¾¹¾Ç²ß¼Ò«¬¨Ó¹w´ú¤Úª÷´Ë¯fªº­·ÀI©M¯fµ{

    • ­^¤å¦WºÙ:
    • ­pµe½s¸¹: 112-2221-E-002-188-MY3
    • ¥D«ù¤H: ±i´¼¬P
    • ¸É§U³æ¦ì: °ê¬ì·|
    • ­pµe°õ¦æ´Á¶¡: 2023/8/1 to 2026/7/31
    • ÃöÁäµü: Machine learning, Parkinson disease
    • ºK­n²¤¶:

  4. ¶°µ²´¼¼z¦ç¤Î¼Æ¾ÚÀ³¥Î©ó¹B°Ê´_°·»Pºë·Ç¹B°Ê¬ãµo­pµe

    • ­^¤å¦WºÙ: The Integration of Smart Garments and Data Applications in the fields of Sports Rehabilitation and Precision Exercise
    • ­pµe½s¸¹: NSTC 112-2622-E-002-009
    • ¥D«ù¤H: ±i´¼¬P
    • ¸É§U³æ¦ì: °ê¬ì·|
    • ­pµe°õ¦æ´Á¶¡: 2023/6/1 to 2024/5/31
    • ÃöÁäµü: Machine learning, precision sport
    • ºK­n²¤¶:
      ¥»­pµe³W¹º¨Ã¬ãµo·s«¬ºAªº´¼¼z¦ç¬[ºc¡A´Á±æ¯à°÷Âǥѳзs¬[ºc¦Ó¤j´T­°§C´¼ ¼z¦çªº¦¨¥»¡A¨Ã´£¨Ñ§óµÎ¾A¤Î¤è«K¥i¦æªºÅéÅç¡A¶i¦Ó¼W¥[®ø¶OªÌ¨Ï¥Îªº·NÄ@¡CµM¦Ó °£¤F°ò¥»ªº¥Í²z¼Æ­È¦¬¶°¦p¤ß«ß¡B©I§lª¬ªpµ¥¥~¡A¨­Åé°Ê§@ªººë·Ç«×¡A¬O»`¶°¨­Åé ¸ê°T§@¬°°ª¶¥À³¥Î¤ÀªRªº¤@¤j­«ÂI¡C§Q¥Î´¼¼z¦ç¤W·P´ú¾¹©Ò¦¬¶°¨ìªº°T®§¡A¨Ã»²¥H ¨Ï¥Î¾÷¾¹¾Ç²ß¡]¨Ò¦p²`«×¯«¸gºô¸ô¡^¨Ó¶i¦æ2D°©Àf¬[ºc¨ì3D¦ì¸mªº¹w´ú¡A¨Ï¥»¨t ²Î¯à°÷·Ç½T¦a¹w´úÃö¸`ÂI¦b3Dªº¦ì¸m¤Î­y¸ñ¡A¨Ã¥Ñ¦¹3D¤¤ªº¦ì¸m©Î­y¸ñ¨Ó§PÂ_¤Î ®Õ¥¿­Ó¤Hªº¦U¶µ¹B°Ê°Ê§@¡A¥H«K¥Î©ó¹B°Ê°V½m¡BÂåÀø´_°·¡B°·±dºÞ²z¡B°ªÄÖ·ÓÅU¡B ºò«æ¨¾Å@¡B¹CÀ¸µ¥°ª¶¥À³¥Î»â°ì¡C

  5. ¶}µoÁ{§É§Y®ÉÀË´ú»P¹w«á¨t²Î©óºC©ÊµÇŦ¯fªººë·Ç¶EÂ_»PªvÀø(2/3)

    • ­^¤å¦WºÙ:
    • ­pµe½s¸¹: 112-2218-E-002-055-
    • ¥D«ù¤H: ±i´¼¬P
    • ¸É§U³æ¦ì: °ê¬ì·|
    • ­pµe°õ¦æ´Á¶¡: 2022/8/1 to 2025/7/31
    • ÃöÁäµü:
    • ºK­n²¤¶:

  6. AIÁn¾Ç½Õ­µ»Pµû´ú¨t²Î

    • ­^¤å¦WºÙ: AI Acoustic Evaluation System
    • ­pµe½s¸¹:
    • ¥D«ù¤H: ±i´¼¬P
    • ¸É§U³æ¦ì: µØºÓ¹q¸£
    • ­pµe°õ¦æ´Á¶¡: 2021/11/1 to 2022/10/31
    • ÃöÁäµü: Machine learning, sound quality assessment¡B, d
    • ºK­n²¤¶:
      ¦b¨Ï¥Î´¼¼z«¬¤â¾÷µ¥¸Ë¸m¤¤ªº­µ¼Ö¼½©ñµ¥À³¥Î®É¡A­µ½è¬O¨Ï¥ÎªÌ·|«D±`Ãö¤ßªº°ÝÃD¤§¤@¡C¨Ò¦p¥Ø«e¦b¦h®a¦ê¬y­µ¼Ö«~µP¡A¦pKKBOX¤ÎSpotifyµ¥¤½¥q¬ÛÄ~´£¨Ñ°ª­µ½è¡A¤D¦ÜµL·l­µ½è¤§­µ·½ªº±¡ªp¤U¡A¤£¦P¼tµP¤§´¼¼z«¬¤â¾÷ªºªº¼½©ñ·l¯Óµ{«×¡A±N¦¨¬°¨Ï¥ÎªÌ¬O§_ÁʶR¨ä¼tµP©Ò±À¥Xªº³]³Æªº­«­n¨Ì¾Ú¡C¦]¦¹¡A­µ½èªº«ÈÆ[µû¶q¡A±N¦¨¬°´¼¼z«¬¤â¾÷»s³y°Ó¤º³¡¹ï²£«~©w¦ì¤ÎÄvª§¤Oµû¦ôªº­«­n«ü¼Ð¡C¥»­pµe¹w­pµ²¦X°T¸¹³B²zºtºâªk»P²`«×¾Ç²ßµ¥¤èªk¡A¨Ã³z¹L¤j¶q¾Ç²ß°ª«ü¼Ð©ÊªºÄv«~¦piPhone¡A¬Æ¦Ü¬O¸ó¬Éªº°ª¶¥­µÅT©Î¦Õ¾÷ªº¿é¥X­µ·½¤§©Ê½è¡A¥HÂǦ¹³]­p¥X¯à±N¥DÆ[­µ½è·P¨üÂର«ÈÆ[¼Æ¾Úªºµû´ú¤èªk¡C

  7. »y­µ¿ëÃѼҲն}µo

    • ­^¤å¦WºÙ: Speech Recognition for Chinese News from Mainland
    • ­pµe½s¸¹:
    • ¥D«ù¤H: ±i´¼¬P
    • ¸É§U³æ¦ì: °ê®a°ª³t¹q¸£¤¤¤ß
    • ­pµe°õ¦æ´Á¶¡: 2021/10/1 to 2022/9/30
    • ÃöÁäµü: Automatic speech recognition, speech to texts for news
    • ºK­n²¤¶:

  8. ¾A¥Î©ó´¼¼zªA°Èªº¥i«H¿àAI¥ý¶i§Þ³N¬ã¨s¡G¡]¤l­pµe¤@¡^¨ã¥i«H«×»P¥i¾a«×ªº¦¸¥@¥NAI«ÈªA

    • ­^¤å¦WºÙ: Advanced Technologies for Designing Trustable AI Services
    • ­pµe½s¸¹:
    • ¥D«ù¤H: ±i´¼¬P
    • ¸É§U³æ¦ì: °ê¬ì·|
    • ­pµe°õ¦æ´Á¶¡: 2021/10/1 to 2025/9/30
    • ÃöÁäµü:
    • ºK­n²¤¶:
      ¥»¬ã¨s¸s­pµe¨Ì´`¥xÆW2030¬ì§ÞÄ@´º¤è¦V¡X³Ð·s¡B¥]®e¡B¥ÃÄò¡A¥H¼Ö¬¡ªÀ·|¤Î´¶ ´f¬ì§Þ¬°¥D­n¤è¦V¡A¬ã¨s¶}µo¥i«H¿àAI§Þ³N¡A¾É¤J¥¼¨Óªº´¼¼zªA°È¡A¹ê²{´¶´f¬ì§Þ ©Ò«üªºµo®i¥H¤H¬°¥»ªº´¼¼z¬ì§Þ¡A±j¤Æ·s¿³¬ì§Þªºªv²z»PÀ³¥Î¡C¥»¬ã¨s¸s­pµe¬°³æ ¤@¾ã¦X«¬­pµe¡A§tÁ`­pµe»P7­Ó¤l­pµe¡A¥Ñ¥x¤j¡B²M¤j¡B¥H¤Î¤¤¬ã°|AI§Þ³N»â°ì³» ¦y¹Î¶¤²Õ¦¨¡A³¯«H§Æ±Ð±Â¬°Á`­pµe¥D«ù¤H¡A¤À¦¨3­Ó¬ã¨s¸s¡G¾÷¾¹¾Ç²ß(¤l­pµe¤@¡B ¤G)¡B¦h´CÅé³B²z(¤l­pµe¤T¡B¥|)¡B¥H¤Î¤H¥»­pºâ(¤l­pµe¤­¡B¤»¡B¤C)¡C¥»­pµe¤§¬D ¾Ô¥D¶b¬°¡i¾A¥Î©ó´¼¼zªA°Èªº¥i«H¿à¤H¤u´¼¼z¥ý¶i§Þ³N¡j¡A±N±q¤£¦P§Þ³N¨¤«×¡B¤£ ¦P¸ê®Æ«¬ºA¡B¥H¤Î¦U­Ó¶}µo»P¨Ï¥Î¶¥¬q¡A¶i¦æ¥i«H¿àAIªº«e¤¬ã¨s¡C¥Ñ¸ê®Æ¦¬¶°¡B §Þ³N¶}µo¡B¦ÜÀ³¥Î¸¨¦a¡A±´°Q³y¦¨¤H¤u´¼¼z/¾÷¾¹¾Ç²ß§Þ³N«H¿à°ÝÃDªº¦]¯À¡A¥]¬A °¾»~¡B¤½¥­©Ê¡B³z©ú©Ê¡B¥i¸ÑÄÀ©Ê¡B¥i¦^·¹©Êµ¥¡C¦U¤l­pµe±N¥H¯S©wªº´¼¼zªA°ÈÀ³ ¥Î¡AÅé²{©Ò¶}µo§Þ³Nªº®Ä¯q»P¹ê»ÚÀ³¥Î¡CÁ`­pµe¥Ñ¬ì§Þ³¡¸É§U»O¤j¤H¤u´¼¼z§Þ³Nº[ ¥þ´T°·±d·ÓÅ@Áp¦X¬ã¨s¤¤¤ß¹Î¶¤­t³dºÞ²z¡B³W¹º¡B¤Î°õ¦æ¡CÁ`­pµe±N¾ã¦X¦U¤l­pµe ªº¬ã¨s¦¨ªG¡A´£¥X¥iªx¤Æ¹B¥Î©ó´¼¼zªA°È²£·~ªº¨ã°ª«×¥i«H¿àªº¤H¤u´¼¼z§Þ³N¾ãÅé ¤è®×¡C

  9. ºqÁn¤ÀÂ÷ DNN model À³¥Î©ó TV SOC ¥­¥x

    • ­^¤å¦WºÙ: Singing Voice Separation via DNN Models on TV SOC Platform
    • ­pµe½s¸¹:
    • ¥D«ù¤H: ±i´¼¬P
    • ¸É§U³æ¦ì: ·ç¬R¬ì§Þ
    • ­pµe°õ¦æ´Á¶¡: 2021/6/1 to 2021/5/31
    • ÃöÁäµü: Singing voice separation, embedded systems, AI
    • ºK­n²¤¶:

  10. ¦Û°Ê¤Æ­Ó¤H¹q¸£«~½è±±ºÞ´ú¸Õ¨t²Î

    • ­^¤å¦WºÙ: Automatic PC quality control and testing systems
    • ­pµe½s¸¹:
    • ¥D«ù¤H: ±i´¼¬P
    • ¸É§U³æ¦ì: ©MºÓÁp¦X¬ì§Þ
    • ­pµe°õ¦æ´Á¶¡: 2020/11/1 to 2021/10/31
    • ÃöÁäµü: Machine learning, pattern recognition, automatic testing
    • ºK­n²¤¶:
      ¥»­pµe±ý¸Ñ¨M©MºÓ¤º³¡´ú¸Õ¬yµ{ªº¦Û°Ê¤Æ°ÝÃD¡C¦b©MºÓ¤º³¡´ú¸Õ¬yµ{¤¤¡A´ú¸Õ¤uµ{®v¥²¶·¤£Â_¹ï±ý´ú¸Õªº¾÷¥x¶i¦æ¦UºØ¦U¼Ëªº´ú¸Õ¡A¥ÑBIOSºÝ¡B¶}¾÷»y¨¥´ú¸Õ¡B¦U¦¡¨t²Î´ú¸Õ(¥]§tÅX°Êµ{¦¡¡B»y¨¥¡BÅã¥d¶]¤Àµ¥)¡A¥Ñ©óºØÃþÁc¦h¡A¬yµ{½ÆÂø¥B¯Ó¶O¤H¤O¡A¦]¦¹©MºÓ¤º³¡«æ»Ý²Î¾ã¤@ºØ¤è¦¡¡A¥H¸Ñ¨M´ú¸Õ¬yµ{¦Û°Ê¤Æ°ÝÃD¡A¥H´Á¯àÄÀ©ñ¤H¤O¨Ó°µ§ó¦³®Ä¯àªº²£¥X¡C

  11. «Ø¥ß¤H¤u´¼¼z¦h­±¦V¤ÀªR¥­¥x¥H¹w´ú¡B¶EÂ_¤Î¤¶¤J¤Úª÷´Ë¤ó¯g»P¬ÛÃö¯«¸g°h¤Æ©Ê¯e¯f

    • ­^¤å¦WºÙ: Developing an artificial intelligence-based multidomain sensing platform to predict and early intervene Parkinson's disease in aging society
    • ­pµe½s¸¹: MOST 109-2221-E-002 -163 -MY3
    • ¥D«ù¤H: ±i´¼¬P
    • ¸É§U³æ¦ì: ¬ì§Þ³¡
    • ­pµe°õ¦æ´Á¶¡: 2020/8/1 to 2021/7/31
    • ÃöÁäµü: Neurodegenerative disorders; Parkinson's disease; Dementia; Artificial Intelligence; Deep Learning; Machine Learning; Facial recognition; Speech recognition; Human Pose Estimation
    • ºK­n²¤¶:
      ¥@¬É¤H¤f°ªÄ֤ơA¤Úª÷´Ë¤ó¯g¬°³Ì±`¨£ªº¯«¸g°h¤Æ©Ê¯e¯f¤§¤@¡C¨å«¬¯gª¬¬°¤£¦Û¥D Ÿ§Ý¡BªÏÅé°Ê§@½wºC»P¨B¥ï¤p¸H¨B¡A«á´Á©ö­«¤ß¤£Ã­®e©ö¶^­Ë¡A¨Ã¦X¨Ö¦X¨Ö»{ª¾¥\ ¯à¤U­°¬Æ¦ÜºtÅܬ°¤Úª÷´Ë¯g¥¢´¼¯g¡A¯f²zÅܤƬ°¦h¤ÚÓi¯«¸g²Ó­M¤¤¥X²{a- synuclein¤£¥¿ªøªº¨I¿n¾É­P¯«¸g²Ó­M­ä¤`¡A¥Ø«eµLªk®Úªv¡A¹ï®a®xªÀ·|­t¾á·¥¤j ¡CÁ{§É¤W¡AÁÙ¦³¤@¸sÃþ¦ü¤Úª÷´Ë¤ó¯g¡A¦ý¦X¨Ö¨ä¥L¨t²Î°h¤Æ(¨Ò¦p¦Û¥D¯«¸g¡B¤p¸£ ¨t²Î¡BÃBù®¸­¥Ö½èµ¥)ªº«D¨å«¬¤Úª÷´Ë¯g­Ô¸s±wªÌ¡A³o¸s¯g­Ô¸s±wªÌ¹ïÃĪ«ªvÀø¤Ï À³»P¹w«á¤£¨Î¡C¥»¹Î¶¤¤§¥x¤jÂå°|¯«¸g³¡¦h¦~«e¤w§Q¥ÎºÏ´î¶q§K¬ÌÀË´ú§Þ³N¡A¿W¨B »â¥ý¥þ²y¹F¨ì¦å²GÀË´ú¤Úª÷´Ë¤ó¯g¦å²G¥Íª«¼Ð»xªº¶¥¬q¡A¥H¨ó§UŲ§O¤Úª÷´Ë¤ó¯g»P ¤Úª÷´Ë¯g­Ô¸s¡C§Ú­Ì¦b¹L¥h¨â¦~¤w©Û¶Ò97¦ì65·³¥H¤W¤§°·±dªøªÌ»P206¦ì¤£¦PÄY­« µ{«×¤§¤Úª÷´Ë¯g±wªÌ¡A§¹¦¨¦å²G¤¤¤À¤l­P¯f³J¥Õ¼Ð°O»P¸£³¡µ²ºc©ÊºÏ®¶¼v¹³¼Æ¾Ú¡C °£¦¹¤§¥~¡A¦b¤Úª÷´Ë¯gª¬¥X²{¤§«e¡A¯f±w¨ä¹ê¤w¶}©l¥X²{ªí±¡´î¤Ö¡B»¡¸ÜÁn­µÅܤp ¡B¦r¸ñ§ïÅÜ»P¨BºAÅܺC©ì¨Bµ¥²Ó·L¼x¥ü¡C¦³Å²©ó¥»¹Î¶¤¹L¥h²`¯ÑÁy³¡ªí±¡¡B»y­µ¿ë ÃѤλy½Õ¬Û¦ü«×¡A¨Ã¤w«Ø¥ßªÏÅé°Ê§@°»´ú¼Ò¦¡¡A¥»­pµe±Nµ²¦XÁ{§É»P²`«×¾Ç²ß¸ê°T ¤uµ{¶i¤@¨B¾ã¦X¤W­z¸£³¡³¡¡B¦å²G»PÁ{§Éªí¼xµ¥¼Ð»x¡A¹w­p¦¬®×»Pªø´Á°lÂÜ100¦W °·±dªøªÌ¡B200¦W¤£¦PÄY­«µ{«×ªº¤Úª÷´Ë¯g±wªÌ»P100¦W«D¨å«¬¤Úª÷´Ë¯g­Ô¸s±wªÌ ¡A¥»­pµe¥Øªº¦p¤U:
      1. ¾ã¦X¾îÂ_­±(cross-sectional)¤§°·±dªøªÌ¡B¤£¦PÄY­«µ{«×¤§¤Úª÷´Ë¯g±wªÌ»P¤Úª÷ ´Ë¯g«á¸s¯f±w¤§¸£¼v¹³»P¦å²G¥Íª«¼Ð°O¸ê°T¡A¥H²`«×¾Ç²ßºtºâªk¡A«Øºc¥i¥HŲ§O¤£ ¦P¯e¯f±Ú¸s»P¯fµ{ÄY­«µ{«×¤§¼Ò¦¡¡C
      2. ªø´Á°lÂÜ(prospective cohort)°·±d¦Ñ¦~ªøªÌ»P¤£¦PÄY­«µ{«×¤§¤Úª÷´Ë¤ó¯g±wªÌ ¡AÂǥѵ²¦X­±³¡ªí±¡¤ÀªR¡B»y­µ¤ÀªR¡BªÏÅé°Ê§@¡B¨BºA¤ÀªR»PÁ{§É¯«¸g¤ß²zµû¦ô ¡A¥H¤H¤u´¼¼z¹Bºâªk«h¨Ó¤ÀªR³o¨Ç¦h­±¦Vªº¸ê®Æ¡A¹w´ú¿©±w­·ÀI»P¯fµ{°h¤Æªº­Ó§O ®t²§©Ê¡A¥H°µ¬°¨ó§U¯e¯f©óÁ{§É«e´Áªº¶EÂ_»P¶q¤Æ¯«¸g°h¤Æªº³t«×¤§¾ã¦X¦¡¤ÀªR¥­ ¥x¡C
      3. ÂǥѤW­z¥Øªº¤@»P¤G§ä¥X¤§Á{§É«e´Á±wªÌ¡A§Ú­Ì±N¶}µoµêÀÀ¹ê¹Òªº¤¬°Ê¹CÀ¸¦¡ °V½m¡A¦A¥H¥Ø¼Ð¤G«Øºc¤§¦h­±¦V¤ÀªR¥­¥x¤ÀªR¬ÛÃö«ü¼Ð¶i¨B¤§´T«×¡A¥H°µ¬°¦­´Á¤¶ ¤J´î½w°Ê§@¥\¯à»P»{ª¾¥\¯à°h¤Æ¤§µ¦²¤¡C
      ¥»­pµeªºµ²ªG¡A±N¦³§U©ó«Ø¥ß¹w´ú¤Úª÷´Ë¤ó¯g¿©¯f­·ÀI»P¯fµ{´c¤Æ¤§¤H¤u´¼¼z¼Ò¦¡ ¡A¤Î¦­°»´ú°·±dªøªÌÂàÅܬ°¤Úª÷´Ë¤ó¯g©Î¬O¯f±¡Âà§é´c¤Æ¤§¶Àª÷ÃöÁä®É¶¡ÂI¡A±N¥i ¨ó§U¥¼¨Ó¯«¸g«OÅ@µ¦²¤©ÎÃĪ«¤§»²§U¤ÀªR¥­¥x¡C

  12. ¾A¥Î©óÃä½t¹Bºâªº³ê¿ôµü¿ëÃѨt²Î

    • ­^¤å¦WºÙ: Keyword spotting on edge devices
    • ­pµe½s¸¹:
    • ¥D«ù¤H: ±i´¼¬P
    • ¸É§U³æ¦ì: Ápµú¬ì§Þ
    • ­pµe°õ¦æ´Á¶¡: 2020/6/1 to 2021/5/31
    • ÃöÁäµü: Keyword spotting, edge computing, speech recognition, deep neural networks
    • ºK­n²¤¶:
      Google©ó2010¦~´£¥Xªº"Google Search by Voice", ¬°ªñ¤Q¦~ªº»y­µ©R¥O§Þ³N¥ß¤U¤Fí©Tªº³þ°ò, ¨Ï±o¦bAndroid/iOS¸Ë¸m¤Î´¼¼z­µ½c¤W³£¯à¦³¦¨¼ôªº»y­µ§U²zªA°È, ¨Ò¦pGoogle Assistant¡BApple Siri¡BAmazon Alexaµ¥. ÀHµÛ»y­µ¿ëÃѧ޳Nªº¶i¨B, »y­µ§U²zªA°ÈªºÀ³¥Î½d³ò·U¨Ó·U¼sªx, ¥]¬A¨®¸ü¨t²Î¥H¤Î´¼¼z®a¹qªº»y­µ¤¬°Ê. ¥»­pµe±N¥H´¼¼z®a®x®Ö¤ßªº´¼¼z¹qµø¬°¥D­n¼Ðªº, ¬°®a®x¦¨­û´£¨Ñ»y­µ§U²zªA°È¡A¨Ò¦p¶}±Ò¡BÃö³¬¹qµø, ´£°ª¡B­°§C­µ¶q, ¬d¸ß¤Ñ®ð¡B®É¨èªíµ¥ªA°È¡C

  13. ´Ó°ò©ó¹q¸£µøı¤§µL¤H¾÷Á×»Ù¬ã¨s­pµe

    • ­^¤å¦WºÙ: UAV obstacle avoidance system with computer vision
    • ­pµe½s¸¹:
    • ¥D«ù¤H: ±i´¼¬P
    • ¸É§U³æ¦ì: ¤¤¬ì°|
    • ­pµe°õ¦æ´Á¶¡: 2020/1/1 to 2020/12/31
    • ÃöÁäµü: Unmanned aerial vehicle, drone, UAV, simultaneous localization and mapping, SLAM, military application, computer vision, monocular camera, surveillance, obstacle avoidance, convolutional neural networks, CNN, Inspect/RestNet, deep learning, GPU computing, parallel computing, Virtual reality, augmented reality, detection, reconnaissance, positioning, self-driving technology, edge computation, artificial intelligence
    • ºK­n²¤¶:
      ªñ¦~¨ÓµL¤H¾÷¬ÛÃöÀ³¥Î³vº¥¦h¤¸¦Ó¼sªx¡A¦b­x¨Æ¥Î³~¤W¡A¤]¦]¨ä»ù·G¡B¤è«K ¡B»´¶qµ¥¯S©Ê¡A³vº¥¦¨¬°­«­n§Þ³N¶}µo­«ÂI¡C±q«Ç¤ºªº©wÂIºÊµøªÅ¶¡¡A¨ì³¥¥~¦Û °Ê©w¦ì»¼°eª««~¡AÁ|¤Z±Ï¨a¡B°»¬d¡B¥H¦Ü©ó§ðÀ»¡A¬Ò¥i¨£µL¤H¾÷ªº¼v¤l¡C¦]¤H¤u ´¼¼z¬ÛÃö§Þ³N¿³°_¡A¦Û¾r§Þ³N¤]¦¨¬°µL¤H¾÷­«ÂI§Þ³NÄvª§³õ°ì¡A¨ä¤¤µL¤H¾÷ªºÁ× »Ù¬°¦Û¾r§Þ³N¤¤ªº°ò¦¡A¦ÓÀ³¥Î©óµL¤H¾÷Á×»Ù¤§¹q¸£µøı§Þ³N¥i¥Î©ó©w¦ì¡BµêÀÀ ¹ê¹Ò(VR)¡BÂX¼W¹ê¹Ò(AR)µ¥¡C §Ú­ÌµÛ²´©óµL¤H¾÷¦Û¾r§Þ³Nªº°ò¦¡A©Ò­n±´°Q©Î¸Ñ¨Mªº°ÝÃD¬°¡GµL¤H¾÷Á×»Ù §Þ³N¡C¥»¬ã¨s´Ó°ò©ó¹q¸£µøı¡A»`¶°»P·Ç³Æ°V½m¸ê®Æ¡A§Q¥Î¨÷¿nºô¸ô (CNN, Convolutional Neural Networks)¬°°ò¦ªºµ²ºc»P²`«×¾Ç²ß¼Ò«¬¡A©Î GoogleªºInception²`«×¾Ç²ßºô¸ô¡A¨Ãµ²¦X¦P¨B©w¦ì»P¦a¹Ïºc«Ø (SLAM, Simultaneous Localization and Mapping)§Þ³N¡A¶È»Ý¥H³æ/ÂùÃèÀYªº¼v ¹³µ²¦XµL¤H¾÷¦ì²¾¡A¦ôºâ¥X»P©P³òÀô¹Òª«Å骺¶ZÂ÷(²`«×)¡A¶i¦Ó°»´ú»Ùê»PÁ׶} »Ùê¡C²{¶¥¬q¡AÃä½t¹Bºâ(Edge Computation)±`¥Î¥H¿é¥X°ªºë·Ç«×ªº±À½×»Pµ²ªG ¡C¥»­pµe±N³z¹LGPUµwÅé¬[ºc¨Ó¶i¦æ¥­¦æ¹Bºâ³B²z¡A¥H´Á¯à§Ö³t¦³®Ä²v¦a±o¨ì·Ç ½Tªº©w¦ì©M«Ø¹Ï¿é¥X¡C ¥»­pµeÀ³¥Î¼h­±ªººc·Q·½¦Û©ó¬ã¨s¶}µoµL¤H¾÷¦Û¾r§Þ³N¡A¦b¥¼¨Ó¥i¥Î©ó¦hºØ ­±¦V¡A¨ä¹ï©ó¥¼¨Óªº¬ã¨s¤u§@¥H¤ÎÀ³¥Î¼h­±¨ã¦³«Ü²`»·ªº¼vÅT©Ê¡C

  14. AI Cup (ºqÁnÂàÃлP©M©¶¿ëÃÑÄvÁÉ)

    • ­^¤å¦WºÙ: AI Cup: Singing Transcription and Music Chord Recognition
    • ­pµe½s¸¹:
    • ¥D«ù¤H: ±i´¼¬P
    • ¸É§U³æ¦ì: ±Ð¨|³¡
    • ­pµe°õ¦æ´Á¶¡: 2019/11/1 to 2021/4/30
    • ÃöÁäµü:
    • ºK­n²¤¶:

  15. ¯¼Â´·~¤u·~4.0AI´¼¼z¶³ºÝ¦â®wº[¬V¦â°t¤è´¼¼zÀ³¥Î

    • ­^¤å¦WºÙ: Textile Industry 4.0: AI Intelligent Color Library on Cloud and Intelligent Application of Dyeing Formulas
    • ­pµe½s¸¹: MOST 108-2745-8-002-007
    • ¥D«ù¤H: ±i´¼¬P
    • ¸É§U³æ¦ì: ¬ì§Þ³¡
    • ­pµe°õ¦æ´Á¶¡: 2019/11/1 to 2020/10/31
    • ÃöÁäµü:
    • ºK­n²¤¶:

  16. ¤H¤u´¼¼z©óª÷¿Ä¬ì§Þ¤§À³¥Î

    • ­^¤å¦WºÙ: Applications of AI over FinTech
    • ­pµe½s¸¹:
    • ¥D«ù¤H: ±i´¼¬P
    • ¸É§U³æ¦ì: ¥É¤s»È¦æ
    • ­pµe°õ¦æ´Á¶¡: 2019/6/1 to 2021/5/31
    • ÃöÁäµü:
    • ºK­n²¤¶:

  17. S3: ¥þ¤è¦ì­µ¼Ö®T¼Ö¾Ç²ß¥­¥x (2/2)

    • ­^¤å¦WºÙ: S3: A Universal Music Platform
    • ­pµe½s¸¹: MOST 107-2823-8-002-009
    • ¥D«ù¤H: ±i´¼¬P
    • ¸É§U³æ¦ì: ¬ì§Þ³¡
    • ­pµe°õ¦æ´Á¶¡: 2018/12/1 to 2019/8/30
    • ÃöÁäµü:
    • ºK­n²¤¶:
      ¥»­pµe§Æ±æ³z¹L¥»¹êÅç«Ç¦b­µ¼Ö¤ÀªR»PÀ˯Á¤G¤Q¦~¨Ó²`¯Ñªº§Þ³N¯à¶q¡A³Ð³y¥X¥þ¤è¦ì­µ¼Ö®T¼Ö¾Ç²ß¥­¥x©Ò»Ý¤§¬ÛÃö§Þ³N»P°ò¦³]¬I¡A¬°¥xÆW¸gÀÙ¶}ÅP¥H¥þ²y¥«³õ¬°¶D¨Dªº¬¡¤ô·½ÀY¡C¥þ²yªº­µ¼Ö¥«³õ²£­ÈÃe¤j¡A¦b¥d°ÕOK¤è­±¡ASmule Sing¦b¥þ²yªº¤U¸ü¶q¤w¸g¶W¹L¤@»õ¡A±M¥Î©ó¥d°ÕOKªºµL½uÂŪ޳Á§J­·¦b¼Ú¬ü¤]¦³1.32»õ¬ü¤¸ªº¥«³õ¦ô­È¡F¦b­µ¼Ö±Ð¨|¤W¥þ²y¥ç¦³45»õ¬ü¤¸ªº¥«³õ¡C¦b¥»­pµe¤¤¡A§Ú­Ì±N¥HAI¤Î¾÷¾¹¾Ç²ß¬°°ò©³¡A¶}µo¬ÛÃö²£«~©Ò»Ýªº¦U¶µÃöÁä§Þ³N¡A¥]§t¥D°Ê¾¸­µ®ø°£¡]¥i¥Î©ó¦øªA¾¹ºÝ©M«È¤áºÝªº¸Ë¸m¡^¡B³æÁn¹D­µ·½¤ÀÂ÷¡]¥i¥Î©ó­µ°T­µ¼Ö¤Î»y­µ¡^¡B¤HÁyªí±¡¿ëÃÑ¡BºqÁn»P¦ñ«µ¦P¨B¡B­µ°ª§ïÅܤΤHÁn®ø°£ªºµwÅé¹ê²{¡BºqÁn¬ü¤Æµ¥¡A³o¨ÇÃöÁä§Þ³N¥i¥H¨Ï¥Î©óB2Bªº°Ó·~¼Ò¦¡¡]³o¬Oµû¼f©e­û±j½Õªº­«ÂI¡^¡A¥i¥H¥Î©ó¦U¶µ°Ó·~ªA°È»PÀ³¥Î¡A¨Ò¦pµø°T­«»s¨t²Î¡]¯à°÷©â¥X¨Ï¥ÎªÌªº»y­µ¨Ó¶i¦æ¶i¤@¨Bªº³B²z¡^¡B½u¤W¥d©ÔOK¡]¥i¥H¤ä´©ºq°Ûµû¤À¡B¥i¥H¨Ï¥ÎYoutubeªº­µ¼Ö¡^¡B³Á§J­·¦¡¥d©ÔOK¡]¯à°÷¹ï¤H¥Í¤Î­µ¼Ö¤É­°key¡B¨Ï¥ÎYoutube­µ¼Ö¡^¡B­µ¼ÖÃý«ß¹CÀ¸¡]¹CÀ¸ÃÐ¥i¥H¦Û°Ê²£¥Í¡B¥i¥H¨Ï¥Î Youtube ªº­µ¼Ö¡^¡B¹q¸£»²§U­µ¼Ö¾Ç²ß¤u¨ã¡]¨ã¦³¦Û°ÊÃСB¦Û°Ê½­¶¡B¦Û°Êµû¤Àµ¥¥\¯à¡^¡B­µ¼ÖÀ˯Á¨t²Î¡]¨Ï¥Î­ó°Û¡B»y­µ©Î­ì¥Í­µ¼Ö¤ù¬qµ¥¡^¡B»y­µ¼W±j¡]¨Ò¦p¥Î©ó¨®½ø¤º³¡ªº»y­µ¿ëÃÑ¡^µ¥¡C¦¹¥~¡A§Ú­Ì¤]·|ªá¤Ö³¡¤À¸ê·½©óB2Cªº°Ó·~¼Ò¦¡¡A§Q¥Îºë·ÇºqÁnµû¤À»P¦P¨B§Y®É¾¸Án®ø°£¡A¶}µo¤â¾÷¥d°ÕOK³nÅéKaraSing¡AÅý¨Ï¥ÎªÌ¦³¹ñ·sªºÅéÅç¡A¹ê²{¯f¬r¦¡¦æ¾P¡A¨Ã³z¹L³Ì·sªº¹CÀ¸¤º®e¦Û°Ê¥Í¦¨»PºVÀ»¿ëÃѪº¥\¯à¡A¹ê§@¤@´Ú·sªº­µ¼Ö¹CÀ¸AutoRhythm¡A¨Ï¥ÎªÌ¯à°÷¨Ï¥ÎYoutubeªº­µ¼Ö¶i¦æºVÀ»¹CÀ¸¡A¸g¥Ñ¹ï­µ¼Öªº¼ô±x©Ê¹F¨ì¹ï¹CÀ¸ªºÂHµÛ«×¡C¦¹¥~¡A§Ú­Ì¥i¨Ï¥Î¤À­y¿ý»sªº­µ¼Ö¨Ó¶}µo­µ¼Ö¾Ç²ß³nÅé¡A¥]§t¦Û°Êµû¤À¤Î¼ÖÃйï¦ìµ¥¥\¯à¡A¹F¨ì´J±Ð©ó¼Öªº¥Ø¼Ð¡CµwÅé¤è­±¡A§Ú­Ì±N¹ê²{¤HÁn¥h°£µ¥¥\¯à¡A¥H´¹¤ù¹ê§@¦bµL½u³Á§J­·¥d©ÔOK¤W¡A©Ô°ª»P¬Û¦ü²£«~ªº®t²§¡C³z¹L»ù³Ð­pµe¤ä´©¡A¥[¤W§Ú­Ìªº§Þ³N¡B³Ð·~¸gÅç¡B»Pª©Åv·~ªÌªº¨}¦nÃö«Y¡A¬Û«H§Ú­Ì¯à°÷²£¥X¤@®a¾ã¦X©Êªº¥­¥x¤½¥q¡A¬°¸gÀÙ°µ¥X¨ãÅé°^Äm¡C

  18. ´Óª«¼v¹³¿ëÃÑ

    • ­^¤å¦WºÙ: Plant Image Recognition
    • ­pµe½s¸¹:
    • ¥D«ù¤H: ±i´¼¬P
    • ¸É§U³æ¦ì: ¤u¬ã°|
    • ­pµe°õ¦æ´Á¶¡: 2018/1/1 to 2018/12/12
    • ÃöÁäµü:
    • ºK­n²¤¶:

  19. µ²¦X¤H¤u´¼¼z»Pª«Ápºô¬ì§Þµo®iºë·ÇºÎ¯vÂå¾Ç¡]¤À¶µ¤@¡Gµo®i©~®aºÎ¯v©I§l¤¤¤î¿zÀ˼ҫ¬¡^

    • ­^¤å¦WºÙ: Developing Artificial Intelligence and IOT Technology on Precision Sleep Medicine
    • ­pµe½s¸¹:
    • ¥D«ù¤H: ±i´¼¬P
    • ¸É§U³æ¦ì: ¬ì§Þ³¡
    • ­pµe°õ¦æ´Á¶¡: 2018/1/1 to 2018/12/31
    • ÃöÁäµü:
    • ºK­n²¤¶:
      ¥»¬ã¨s§Æ¾¬³z¹LÁn­µ§@¬°¥D­n±´°Qªº¦]¤l¡A¦b¹L¥h¬ã¨sªº¤W¡A¨ÃµL°w¹ïÁn­µ¨Ó¶i¦æºÎ¯v«~½èªº¤ÀÃþ¡C¦]¦¹¡A¥»¬ã¨s¥D­nªº¹ï¶H¬°°w¹ïBMI¬°27¥H¤U«DªÎ­D«¬ªº¯f±w¡A³z¹LPSGªºÀË´ú¡B§C©I§l«ü¼Æ(Apnea-Hypopnea Index, AHI)¡B©~®a¬ïÀ¹¦¡¸Ë¸mªºÀË´ú¼Æ¾Ú¡B°Ý¨÷ªº¶ñµªµ²ªG¡A¥ÑÂå¥Í±M®a¨Ó§P©wªººÎ¯v«~½è¦nÃa¡A¨Ãµ¹©wºÎ¯v¨ü¼vÅTÄY­«µ{«×ªº¤À¯Å¡C¤§«á¡A¶i¤@¨B°w¹ï¦P¤@¯f±wªºªvÀø«e«á¡A¤£¦P¶¥¬qªºÁn¯¾¯S½è¶i¦æÁn¯¾¿ëÃѵ²ªG¤ñ¹ï¡A»²¥H²Î­p¼Ò«¬ªº«ÈÆ[¼Æ¾Ú§@¬°¨Ì¾Ú¨ÓÅã¥Ü¸Ó¯f±w©óªvÀø«á¡A¨äÁn¯¾¯S½èªº§ïÅܬO§_Áͪñ¥¼¿©±wOSA¨ü´úªÌªº±`¼Ò«¬ºA¡C¥»¶¥¬q­pµe¥Ø¼Ð¬°¤ÀªR¦U¶µÁn­µ°T¸¹¡A´£¨ÑÁn­µÅܤƪººtºâªk¡A³z¹L²`«×¾Ç²ßºtºâªk¬[ºc¡A¨Ó°V½mOSAÁn¯¾¼Ò«¬¡C¦b°T¸¹¤ÀªR¤À¬°¨â­Ó³¡¤À¤£¦P³õ°ì¡A¤@­Ó¬O¦bÂå°|ºÝ¶i¦æOSA±wªÌªºPSG°T¸¹¦¬¶°¡A¥t¤@­Ó«h¬O¦b©~®aºÝ¶i¦æOSA±wªÌªºÁn­µ¦¬¶°¡C³z¹L¤£¦P³õ°ì©Ò²£¥ÍªºÁn­µ°T¸¹¶i¦æ¤ÀªR¡F­º¥ý¡A³z¹LÁn­µªí²{¦b¥À­µ²M´·«×ªº®t²§§@¬°¯S¼x°Ñ¼ÆÂ^¨úªº¨Ì¾Ú¡A¨Ó¶i¦æºÎ¯v«~½è¨ü¼vÅTµ{«×¤j¤pªº¿ëÃÑ¡B¤ÀÃþ¡A¨ä¦¸¡A¥i¥H³z¹L¦³¨ã¶H»y·Nªºµü»y¹Ï¥d¡A½Ð¯f±w©óºÎ«e¡BºÎ«á½Æ»w®ÔŪ¡A´ú¶qµü»y¿ë»{¤ÏÀ³®É¶¡ªº®t²§¬O§_¼vÅT»{ª¾¡C¦A«h¡A¬O¨Ï¥Î¶K¤ù¦¡³Á§J­·©ó¯f±w³ïÄVªþªñ¡A³z¹L¤@©]ºÎ¯v¹Lµ{¦¬¶°¸û¬°²M´·ªºÂMÁn­µ½è¡A»PPSG°T¸¹©M¬ïÀ¹¦¡¸Ë¸mªº¤ß²vÅܲ§«×(HRV)¶i¦æ¤ñ¹ï¡A¥Î©ó§PÂ_¨ä±¡ºüªº¥¿¦V©Î­t¦V¼vÅT¡C

  20. »O¤jÂ寫¡Vºë·ÇÂåÀø¤H¤u´¼¼z»²§U¨Mµ¦¨t²Î¡m¤l­pµe¥|¡Gºë·ÇÂåÀø¤H¤u´¼¼z¶}µo¡^

    • ­^¤å¦WºÙ: NTU Medical Genie ¡V AI Decision Support System for Precision Medicine (Subproject 4: AI Technologies for Precision Medicine)
    • ­pµe½s¸¹: MOST 107-2634-F-002-015
    • ¥D«ù¤H: ±i´¼¬P
    • ¸É§U³æ¦ì: ¬ì§Þ³¡
    • ­pµe°õ¦æ´Á¶¡: 2018/1/1 to 2018/12/31
    • ÃöÁäµü:
    • ºK­n²¤¶:
      ¦b²{¤µªº¸ê°T§Þ³N»P¸ê®Æ°T®§§Ö³t¿±µÈªº®É¥N¡A¦UºØ§Þ³Nªº³Ð·s»P´¼¯à¤Æ³£±N¬°ÂåÀø¤H­û´£¨Ñ§óºë·ÇªºÂå¾Ç¶EÂ_»PªvÀø¡A¨Ã¥BµÛ­«¦b­Ó¤H¤Æªº¯e¯f¹w¨¾¡Bµû¦ô¶EÂ_¡BªvÀø©M±d´_·ÓÅ@­p¹º¡A¬O¥Ø«eÁ{§É¤W±`¨£ªº­«­nijÃD»P§x¹Ò¡AµM¦Ó¡AÂåÀø¤H­û«æ»Ý²{¥N¤Æ¸ê®Æ³q°T§Þ³Nªº»²§U¡A°w¹ï¦U¦¡¦U¼ËªºÂå¾Ç¸ê°T«H®§¶i¦æ·J¾ã»PÂk¯Ç¡C¬°¤F¦³®Ä¸Ñ¨M¤W­zijÃD¡A¥[³t­Ó¤H¤ÆÂåÀø»PÂàĶÂå¾Çªº¾ã¦X¡A¥»­pµe±N°w¹ï©Ò¦³ÂåÀø¸ê®Æ¶i¦æ¾ã¦X¡A¥]§tªù«æ¶E©M¦í°|ªº¶EÂ_¡B¥ÎÃÄ¡B³B¸m¡B¥Í¤ÆÀËÅç³ø§i¡BÂå¾Ç¼v¹³³ø§i¡B°ò¦]¡B®a±Ú¯f¥v¡B­Ó¤H¥Í¬¡«¬ºA¥H¤Î¤½¶}ªºªÀ·|Àô¹Ò»P°·±dÃö«Yµ¥¬ÛÃö¸ê®Æ¡A¥H«Øºc­Ó¤H¤Æªº§¹¾ãÂåÀø¸ê®Æ¡A¨Ã¹B¥Î¤j¼Æ¾Ú¤ÀªR¤Î¾÷¾¹²`«×¾Ç²ßµ¥²{¥N¤Æ§Þ³N¡A»²¥H·í¥N¹êÃÒÂå¾Ç¤åÄm¤Î¹q¤l¯f¾ú±´°É¡A«Ø¥ß­Ó¤H¤ÆÂåÀø­pµeªººë·ÇÂåÀø»²§U¨Mµ¦¨t²Î¡A´£¨Ñ­Ó¤H¤Æªº¯e¯f¹w¨¾¡B¶Eªv©M±d´_·ÓÅ@ÂåÀø«Øij¡A¨Ã¥H»O¤jÂåÀøÅé¨tÁ{§É¹ê°È±¡¹Ò§@¬°¬ã¨s³õ°ì¡A¬ãµo¨Mµ¦«Øij¤H¤u´¼¼zºtºâªk¼Ò«¬¡A¥H§¹¦¨ºë·ÇÂåÀø»²§U¨Mµ¦¨t²Îªº¶}µo¡C

  21. S3: ¥þ¤è¦ì­µ¼Ö®T¼Ö¾Ç²ß¥­¥x (1/2)

    • ­^¤å¦WºÙ: S3: A Universal Music Platform
    • ­pµe½s¸¹: MOST 106-3114-8-002-003
    • ¥D«ù¤H: ±i´¼¬P
    • ¸É§U³æ¦ì: ¬ì§Þ³¡
    • ­pµe°õ¦æ´Á¶¡: 2017/12/1 to 2018/11/30
    • ÃöÁäµü:
    • ºK­n²¤¶:
      ¥»­pµe§Æ±æ³z¹L¥»¹êÅç«Ç¦b­µ¼Ö¤ÀªR»PÀ˯Á¤G¤Q¦~¨Ó²`¯Ñªº§Þ³N¯à¶q¡A³Ð³y¥X¥þ¤è¦ì­µ¼Ö®T¼Ö¾Ç²ß¥­¥x©Ò»Ý¤§¬ÛÃö§Þ³N»P°ò¦³]¬I¡A¬°¥xÆW¸gÀÙ¶}ÅP¥H¥þ²y¥«³õ¬°¶D¨Dªº¬¡¤ô·½ÀY¡C¥þ²yªº­µ¼Ö¥«³õ²£­ÈÃe¤j¡A¦b¥d°ÕOK¤è­±¡ASmule Sing¦b¥þ²yªº¤U¸ü¶q¤w¸g¶W¹L¤@»õ¡A±M¥Î©ó¥d°ÕOKªºµL½uÂŪ޳Á§J­·¦b¼Ú¬ü¤]¦³1.32»õ¬ü¤¸ªº¥«³õ¦ô­È¡F¦b­µ¼Ö±Ð¨|¤W¥þ²y¥ç¦³45»õ¬ü¤¸ªº¥«³õ¡C¦b¥»­pµe¤¤¡A§Ú­Ì±N¥HAI¤Î¾÷¾¹¾Ç²ß¬°°ò©³¡A¶}µo¬ÛÃö²£«~©Ò»Ýªº¦U¶µÃöÁä§Þ³N¡A¥]§t¥D°Ê¾¸­µ®ø°£¡]¥i¥Î©ó¦øªA¾¹ºÝ©M«È¤áºÝªº¸Ë¸m¡^¡B³æÁn¹D­µ·½¤ÀÂ÷¡]¥i¥Î©ó­µ°T­µ¼Ö¤Î»y­µ¡^¡B¤HÁyªí±¡¿ëÃÑ¡BºqÁn»P¦ñ«µ¦P¨B¡B­µ°ª§ïÅܤΤHÁn®ø°£ªºµwÅé¹ê²{¡BºqÁn¬ü¤Æµ¥¡A³o¨ÇÃöÁä§Þ³N¥i¥H¨Ï¥Î©óB2Bªº°Ó·~¼Ò¦¡¡]³o¬Oµû¼f©e­û±j½Õªº­«ÂI¡^¡A¥i¥H¥Î©ó¦U¶µ°Ó·~ªA°È»PÀ³¥Î¡A¨Ò¦pµø°T­«»s¨t²Î¡]¯à°÷©â¥X¨Ï¥ÎªÌªº»y­µ¨Ó¶i¦æ¶i¤@¨Bªº³B²z¡^¡B½u¤W¥d©ÔOK¡]¥i¥H¤ä´©ºq°Ûµû¤À¡B¥i¥H¨Ï¥ÎYoutubeªº­µ¼Ö¡^¡B³Á§J­·¦¡¥d©ÔOK¡]¯à°÷¹ï¤H¥Í¤Î­µ¼Ö¤É­°key¡B¨Ï¥ÎYoutube­µ¼Ö¡^¡B­µ¼ÖÃý«ß¹CÀ¸¡]¹CÀ¸ÃÐ¥i¥H¦Û°Ê²£¥Í¡B¥i¥H¨Ï¥Î Youtube ªº­µ¼Ö¡^¡B¹q¸£»²§U­µ¼Ö¾Ç²ß¤u¨ã¡]¨ã¦³¦Û°ÊÃСB¦Û°Ê½­¶¡B¦Û°Êµû¤Àµ¥¥\¯à¡^¡B­µ¼ÖÀ˯Á¨t²Î¡]¨Ï¥Î­ó°Û¡B»y­µ©Î­ì¥Í­µ¼Ö¤ù¬qµ¥¡^¡B»y­µ¼W±j¡]¨Ò¦p¥Î©ó¨®½ø¤º³¡ªº»y­µ¿ëÃÑ¡^µ¥¡C¦¹¥~¡A§Ú­Ì¤]·|ªá¤Ö³¡¤À¸ê·½©óB2Cªº°Ó·~¼Ò¦¡¡A§Q¥Îºë·ÇºqÁnµû¤À»P¦P¨B§Y®É¾¸Án®ø°£¡A¶}µo¤â¾÷¥d°ÕOK³nÅéKaraSing¡AÅý¨Ï¥ÎªÌ¦³¹ñ·sªºÅéÅç¡A¹ê²{¯f¬r¦¡¦æ¾P¡A¨Ã³z¹L³Ì·sªº¹CÀ¸¤º®e¦Û°Ê¥Í¦¨»PºVÀ»¿ëÃѪº¥\¯à¡A¹ê§@¤@´Ú·sªº­µ¼Ö¹CÀ¸AutoRhythm¡A¨Ï¥ÎªÌ¯à°÷¨Ï¥ÎYoutubeªº­µ¼Ö¶i¦æºVÀ»¹CÀ¸¡A¸g¥Ñ¹ï­µ¼Öªº¼ô±x©Ê¹F¨ì¹ï¹CÀ¸ªºÂHµÛ«×¡C¦¹¥~¡A§Ú­Ì¥i¨Ï¥Î¤À­y¿ý»sªº­µ¼Ö¨Ó¶}µo­µ¼Ö¾Ç²ß³nÅé¡A¥]§t¦Û°Êµû¤À¤Î¼ÖÃйï¦ìµ¥¥\¯à¡A¹F¨ì´J±Ð©ó¼Öªº¥Ø¼Ð¡CµwÅé¤è­±¡A§Ú­Ì±N¹ê²{¤HÁn¥h°£µ¥¥\¯à¡A¥H´¹¤ù¹ê§@¦bµL½u³Á§J­·¥d©ÔOK¤W¡A©Ô°ª»P¬Û¦ü²£«~ªº®t²§¡C³z¹L»ù³Ð­pµe¤ä´©¡A¥[¤W§Ú­Ìªº§Þ³N¡B³Ð·~¸gÅç¡B»Pª©Åv·~ªÌªº¨}¦nÃö«Y¡A¬Û«H§Ú­Ì¯à°÷²£¥X¤@®a¾ã¦X©Êªº¥­¥x¤½¥q¡A¬°¸gÀÙ°µ¥X¨ãÅé°^Äm¡C

  22. À³¥Î²`«×¾Ç²ß¤èªk¤§»yªÌÅçÃÒ§Þ³N

    • ­^¤å¦WºÙ: Speaker Verification Using Deep Learning
    • ­pµe½s¸¹:
    • ¥D«ù¤H: ±i´¼¬P
    • ¸É§U³æ¦ì: ¤¤µØ¹q«H
    • ­pµe°õ¦æ´Á¶¡: 2017/10/20 to 2018/10/19
    • ÃöÁäµü:
    • ºK­n²¤¶:
      ¥»­pµe¥Øªº¬°¬ãµo¤@ºØ»yªÌÅçÃÒªººtºâªk¡A¥H²`«×¾Ç²ß§Þ³Nºë¶i¨ä·Ç½T²v¡A¨Ã¶}µo¥X¤@­ÓÀ³¥Î¦¹ºtºâªk¤§Âú«¬¨t²Î¡C»yªÌ¿ë»{ (Speaker Recognition) ¥D­n®Ú¾Ú¨Ï¥ÎªÌÁn­µªº¯S¼x¡A¿ë§O¨Ï¥ÎªÌªº¨­¤À¡C¦b¤£¦PªºÀ³¥Î¼h­±¤W¡A¤À¬°»yªÌ¿ë§O (Speaker identification) »P»yªÌÅçÃÒ (Speaker verification) ¨â¤jÃþ¡C·í¤¤ªº»yªÌÅçÃÒ§Þ³Nªº¯SÂI¬°¨¾°°©Ê¡B«K§Q©Ê¤Î·Ç½T©Ê¡C¦b¨¾°°©Ê¤è­±¡A¥i³z¹LÀH¾÷«ü©w¨Ï¥ÎªÌ»ÝÁ¿¥X¤§¦r¦ê¡A¨¾¤î¥L¤H°¼¿ýÅѨú¡A´£°ª»yªÌÅçÃÒ¤§¦w¥þ¾÷±K©Ê¡F¦b«K§Q©Ê¤è­±¡A¬Û¸û©ó¶Ç²Î¹q¸Ü¤¤¤ñ¹ï«È¤á­Ó¤H¸ê®ÆªºÅçÃҤ覡¡A»yªÌÅçÃÒ§Þ³N¯à­°§C«È¤áµ¥«Ý®É¶¡¡C¦Ó¦b·Ç½T©Ê¤è­±¡A¥Ñ©ó²`«×¾Ç²ß§Þ³Nªººt¶i¡A»yªÌÅçÃÒ¤§·Ç½T«×¤w¤j´T´£¤É¦Ü½u¤WªA°È¥i±µ¨ü¤§½d³ò¡C¦]¦¹¡A¥»­pµe¥Øªº¬°¬ãµo¤@ºØ»yªÌÅçÃÒªººtºâªk¡A±N¨Ï¥Î²`«×¾Ç²ß (¦p: ²`«×¯«¸gºô¸ô (Deep Neural Network¡ADNN)¡B»¼Âk¯«¸gºô¸ô(Recurrent Neural Network¡ARNN)¡B¨÷¿n¯«¸gºô¸ô(Convolution Neural Network¡ACNN) µ¥) »P¾÷±ñ¾Ç²ßµ¥¬ÛÃö§Þ³N¡A¶}µo¥X¤@­ÓÀ³¥Î¦¹ºtºâªk¤§Âú«¬¨t²Î¡A¹w´Á¦¹¨t²Îªº»yªÌÅçÃҮįà¦b¦X¾AªºÀô¹Ò¤U¯à°÷¹F¨ì¯S²§©Ê(specificity)99.9%¡A¦Ó±Ó·P©Ê(sensitivity)¯à¨ì¹F95%¡A¶i¦Ó¯à°÷¤W½uªA°È¼s¤jªº¤¤µØ¹q«H«È¤á¡C

  23. ´Óª«¼v¹³¿ëÃÑ

    • ­^¤å¦WºÙ: Plant Image Recognition
    • ­pµe½s¸¹:
    • ¥D«ù¤H: ±i´¼¬P
    • ¸É§U³æ¦ì: ¤u¬ã°|
    • ­pµe°õ¦æ´Á¶¡: 2017/9/1 to 2018/8/31
    • ÃöÁäµü: Image recognition, CNN, DNN
    • ºK­n²¤¶:
      ¥»­pµe¥Ø¼Ð¬ãµo¥X¤@®M°ò©óªá»P¸­¤ù¤§´Óª«¿ëÃѨt²Î¡A°w¹ï¤j¦ÛµM¤¤ªº´Óª«¡A§Q¥Î·Ó¬Û¤â¾÷©Ò©çÄ᪺ªá¦·¼v¹³¤Î¸­¤ù¼v¹³¶i¦æ¿ëÃÑ¡C¥D­n¥Øªº¦b©ó¨ó§U¥ÍºA«O¨|±M®a§ó¦³®Ä²v¬d¬Ý´Óª««~ºØ¥H¤Î©w®É¬d¬Ý»P½ñ¬d¦U¦aªºª«ºØ¥HÁA¸Ñ¥ÍºA¥­¿Å©Ê¡A¦]¦¹°Ê¡B´Óª«ªº«~ºØ¿ëÃѹ惡»â°ì±M®a¨Ó»¡¬Û·í­«­n¡C¥»¤èªkªº¬yµ{¹Ï¦p¹Ï1.©Ò¥Ü¡C­º¥ý§Ú­Ì·|¥ý©w¸q´Óª«ª«ºØ¡A¶i¦æ¦Uª«ºØªº¹Ï¤ù·j¶°¨Ã«Ø¥ß¸ê®Æ®w¡A¨Ã¦b»`¶°¨ìªº¹Ï¤ù¤W¼Ð°Oª«ºØªº¯S¼x¡C±µ¤U¨Ó·|¦P®É¶i¦æ¶Ç²Î¾÷±ñ¾Ç²ß¤èªk¥H¤Î²`«×¾Ç²ß¤èªkªº¬ãµo¡C¦b¶Ç²Î¾÷±ñ¾Ç²ß¤èªk¤W¡A·|¥ý¨Ï¥Î¤w¦s¦bªº¯S¼x©â¨ú¤èªk©â¨úª«ºØ¤Wªº­«­n¯S¼x¡A¦A©ñ¨ì¤ÀÃþ¾¹°µ°V½m¡C¦b²`«×¾Ç²ßªº¤èªk¤W¡A·|¨Ï¥Î¨÷¿n¯«¸gºô¸ô (Convolutional Neural Network, CNN) °µ¯S¼x©â¨ú¤Î°V½m¡A¨Ã¦P®É¤ñ¸û¨âºØ¤èªkªºµ²ªG©M§ï¶i¡C

  24. ¨t²Î©Ê¦Û°Ê¤Æ¶K¼Ð

    • ­^¤å¦WºÙ: Systematic Automatic Labeling
    • ­pµe½s¸¹:
    • ¥D«ù¤H: ±i´¼¬P
    • ¸É§U³æ¦ì: ¥É¤s»È¦æ
    • ­pµe°õ¦æ´Á¶¡: 2017/5/1 to 2017/10/31
    • ÃöÁäµü:
    • ºK­n²¤¶:
      ¥»­pµe±N°w¹ïÅU«È¦æ¬°¡Bºô­¶¡BÁʶRª««~¡BPTTµ¥¸ê®Æ¶i¦æ¤ÀªR¡A¯S§O¬O¨Ï¥Î¾÷¾¹¾Ç²ß¨Ó¶i¦æ¦Û°Ê¤Æ¶K¼Ð¡A¥H«K¶i¦æ°w¹ïÅU«È°¾¦nªººë·Ç¦æ¾P¡C

  25. ºë·Ç¦æ¾P»P¨t²Î¤Æ¦Û°Ê¶K¼Ð

    • ­^¤å¦WºÙ: Precision Sales and Systematic & Automatic Labeling
    • ­pµe½s¸¹:
    • ¥D«ù¤H: ±i´¼¬P
    • ¸É§U³æ¦ì: ¥É¤s»È¦æ
    • ­pµe°õ¦æ´Á¶¡: 2017/3/1 to 2017/10/31
    • ÃöÁäµü:
    • ºK­n²¤¶:

  26. ¬F©²¥¨¶q¸ê®Æ¤ÀªR¤u¨ã»P¥­¥x¤l¡G¤l­pµe¤@. ¥¨¶q»y­µ¸ê®Æ¤ÀªR

    • ­^¤å¦WºÙ: Big Speech Data Analytics
    • ­pµe½s¸¹:
    • ¥D«ù¤H: ±i´¼¬P
    • ¸É§U³æ¦ì: ¬ì§Þ³¡
    • ­pµe°õ¦æ´Á¶¡: 2016/7/1 to 2017/6/30
    • ÃöÁäµü:
    • ºK­n²¤¶:
      ¥»­pµe¤§¬ãµo­«ÂI¬°¡u«ÈªA»y­µ¸ê®Æ¤ÀªR¡v¡]speech analytics at call centers¡^¡A³o¬O¤@­Ó¥Ø«e«Ü¼öªùªº¬ã¨s½ÒÃD¡A¥D­n¥Ø¼Ð¬O¸g¥Ñ«ÈªA¤¤¤ßªº¿ý­µ¡A¬ö¿ý«ÈªA¤H­û©MÅU«Èªº¹ï¸Ü¹Lµ{¡A¨Ã¸g¥Ñ¦¹¤j¶q¸ê®Æªº¤ÀªR¡A¨Ó´£°ª«È¤áªA°Èªº®Ä²v¡A¨Ã¼W¶i«È¤á¹ï¦¹ªA°Èªºº¡·N«×¡C

  27. ºq¦±¦±­·¤ÀÃþ¾¹»P±¡¹Ò¤ÀÃþ¾¹¤§¬ã¨s

    • ­^¤å¦WºÙ: Research on Music Genre Classification and Scenario classification
    • ­pµe½s¸¹:
    • ¥D«ù¤H: ±i´¼¬P
    • ¸É§U³æ¦ì: ¤u¬ã°|
    • ­pµe°õ¦æ´Á¶¡: 2016/4/1 to 2016/11/15
    • ÃöÁäµü:
    • ºK­n²¤¶:
      ¥H¾÷¾¹¾Ç²ßºtºâªkµ²¦X­µ°T¯S¼x«Ø¥ßºq¦±¦±­·»P±¡¹Ò¤ÀÃþ¾¹¡A¯àÂǦ¹±j¤Æ½u¤W­µ¼Ö¦ê¬y±ÀÂË¥\¯à¡A¤Þ¾É¨Ï¥ÎªÌ²âÅ¥§ó¦h¤¸¦±­·¡A´£°ª¨Ï¥ÎªÌÂHµÛ«×¡CµM¾÷¾¹¾Ç²ßÀ³¥Î¦b­µ°T¤ÀªR¤Wªº½ÆÂø«×«D±`°ª¡AÂǥѦ¹¦X§@­pµe¡Aµ²¦X¸Ó»â°ì±M®a¾ÇªÌ¡A¥[³t­pµe°õ¦æªº³t«×¡A¨Ã´£°ª¤ÀÃþ¾¹·Ç½T«×¡C

  28. ¥Î©ó­µ¼Ö¸ê°TÀ˯Áªº²`«×¾Ç²ß

    • ­^¤å¦WºÙ: Deep Learning for Music Information Retrieval
    • ­pµe½s¸¹:
    • ¥D«ù¤H: ±i´¼¬P
    • ¸É§U³æ¦ì: ¬ì§Þ³¡
    • ­pµe°õ¦æ´Á¶¡: 2015/8/1 to 2018/7/31
    • ÃöÁäµü:
    • ºK­n²¤¶:
      ÀHµÛ¹q¸£­pºâ¯à¤Oªº§Ö³t±À¶i¡]Âk¥\©ó CPU ¤Î GPU ªº¦@¦P¹B§@¡^¡B°O¾ÐÅ骺¤j¶q¨Ï¥Î¡A¥H¤Î¾÷¾¹¾Ç²ßªº¤èªkºt¶i¡A²`«×¾Ç²ßªº¯«¸gºô¸ô¡]deep-learning neural networks¡^¤w¸g³Q¦¨¥\¦a¥Î¦b¤j¶q¼v¹³¤Î»y­µªº¿ëÃÑ¡A¨ä¥¿½T²v¤w¸g¤j´T¶W¶V¶Ç²Îªº¿ëÃѤèªk¡A¦P®É¤]±È°_¤F¤@ªÑ·sªº¬ã¨s¤ÎÀ³¥Î¼ö¼é¡C¥»­pµe±N±´°Q²`«×¾Ç²ß¦p¦ó¥Î©ó­µ¼Ö¸ê°TÀ˯Áªº¦U¶µ°ò¥»¤u§@¡A¥D­n¥]§t½Æ­µ­µ°T­µ¼Öªº¤HÁn¥D±Û«ß©â¨ú¡]vocal melody extraction from polyphonic audio music¡^¡B¦±­·¤ÀÃþ¡]genre classification¡^¡B±¡ºü¤ÀÃþ¡]mood classification¡^¡B½°Ûºq°»´ú¡]cover song identification¡^¡B­µ°TÁn¯¾¿ëÃÑ¡]audio fingerprinting¡^¡B­ó°Û¿ïºq¡]query by singing/humming¡^¡B¸`©ç°lÂÜ¡]beat tracking¡^µ¥¡C©¹¦~§Ú­Ì°Ñ¥[ MIREX ¤ñÁɪº³o¨Ç¬ÛÃöµû¤ñ¡A³£±o¨ì«Ü³Ç¥Xªº¦¨ÁZ¡A¦ý¬O­Y­n¦Aºë¶i¡A¦ü¥G¦³¤@­Ó¬Á¼þ¤ÑªáªOªùÂe¡A¦b³o­Ó­pµe¤¤¡A§Ú­Ì±N¨Ï¥Î¦UºØ¤£¦Pªº²`«×¾Ç²ß¤è¦¡¡]¥]§t¦UºØ¯«¸gºô¸ôªº¬[ºc¡B¾Ç²ßªk¡BGPU ªº¹ê²{µ¥¡^¡A¹Á¸Õ¬ð¯}³o­Ó¬Á¼þ¤ÑªáªOªùÂe¡A²Ä¤@¦~±N¥H¡u½Æ­µ­µ°T­µ¼Öªº¤HÁn¥D±Û«ß©â¨ú¡v¬°¥D¡F²Ä¤G¦~ªº¥Ø¼Ð«h¬O¡u¦±­·¤Î±¡ºü¤ÀÃþ¡v¡F²Ä¤T¦~ªº¥Ø¼Ð«h¬O¡u­µ°TÁn¯¾¿ë ÃѤν°Ûºq°»´ú¡v¡C

  29. ­µ°T«ü¯¾À˯Á»P¯B¤ô¦L´O¤J§Þ³N

    • ­^¤å¦WºÙ: Audio Fingerprinting & Audio Watermarking
    • ­pµe½s¸¹:
    • ¥D«ù¤H: ±i´¼¬P
    • ¸É§U³æ¦ì: ¤¤µØ¹q«H
    • ­pµe°õ¦æ´Á¶¡: 2015/4/1 to 2016/3/31
    • ÃöÁäµü:
    • ºK­n²¤¶:
      ¥»­p¹º±N¥H­µ°T¯S¼x¿ëÃѧ޳N¡A¥H¤Î­µ°T¯B¤ô¦L´O¤J§Þ³N¡A¹ï²{¦³¼v­µªA°È¤§«~½è¶i¦æ´£¤É¡C¦b­µ°T¯S¼x¿ëÃѧ޳N¤è­±¡A±N³z¹L¯Á¤ÞÂø´ê§Þ³Nªº§ï¨}¡A¨Ã¤Þ¶iGPUµ¥¥­¦æ§Þ³Nªº¤ä´©¡A¥H§Ö³t¦a¹ï¤j¶q´CÅé¸ê®Æ¶i¦æ³B²z¡A¥H´Á¹F¨ì¨t²Î¸ê·½¹B¥Îªº³Ì¨Î¤Æ¡A¥H¤Î­°§C¥Î¤á°e¥X¬d¸ß«áªºµ¥«Ý®É¶¡¡C¹ï©ó­µ°T¯B¤ô¦L´O¤J§Þ³N¡A«h¥i¦b¤H¦ÕµLªk¹îıªº±ø¥ó¤U¡A±N°T®§ÁôÂé󭵰T¤¤¡A¸Ó°T®§¥i¥Î©ó­µ°T¤§¬ÛÃö¸ê°T¡B¼s§i¶Ç¼½¡A¼W¥[­ì­µ°T¤§ªþ¥[»ù­È¡A¥t¤]¥i¥Î©ó¼Æ¦ì¸ê®ÆµÛ§@Åv¤§ºÞ²z¡C

  30. §Ü¤zÂZªº­µ°T¸ê°TÁôÂç޳N

    • ­^¤å¦WºÙ: Noise-robust Information Hiding for Audio Signals
    • ­pµe½s¸¹:
    • ¥D«ù¤H: ±i´¼¬P
    • ¸É§U³æ¦ì: ¸êµ¦·|
    • ­pµe°õ¦æ´Á¶¡: 2015/1/1 to 2015/12/15
    • ÃöÁäµü: ¸ê°TÁôÂáB­µ°T³B²z¡B­µ°T¯B¤ô¦L
    • ºK­n²¤¶:
      ¥»­pµe±N¶}µo§Ü¤zÂZªº­µ°T¸ê°TÁôÂútºâªk¡A¥i±N¯S©w¸ê°T¡]¦p¼v¤ùID¡B²£«~ºô§}µ¥¡^¨Æ¥ý´O¤J­µ°T¤§¤¤¡A¦Ó¥B³o¨ÇÁôÂ꺸ê°T¨Ã¤£·|Åý­µ°T¥¢¯u¡AÅ¥¨ä¨ÓÀ³¸Ó©M­ì¨Óªº­µ°T¤@¼Ë¡C­µ°T¼½©ñ®É¡A¦bªþªñªº¨Ï¥ÎªÌ¥i¥Î¤â¾÷µ¥¤â«ù¦¡¸Ë¸m¦bµu®É¶¡¤ºÅª¥XÁôÂè䤤ªº¸ê°T¡A¨Ò¦pµuºô§}µ¥¡A¨Ã¥i¥ß§Y¦b¤â¾÷¤WÅã¥Ü©Î°õ¦æ¡C¥Ñ©ó©Ò¦³ªº­pºâ³£¦b«e¥x§¹¦¨¡A©Ò¥H§Ú­Ì¥i¥H´î¤Ö«á¥x¦øªA¾¹ªº­t²ü¡A«D±`¾A¦X¤j¶q¨Ï¥ÎªÌ¨Ï¥Î¡C

  31. Layout Sensitivity Model for NTO/CDU APC

    • ­^¤å¦WºÙ: Layout Sensitivity Model for NTO/CDU APC
    • ­pµe½s¸¹:
    • ¥D«ù¤H: ±i´¼¬P
    • ¸É§U³æ¦ì: ¥x¿n¹q
    • ­pµe°õ¦æ´Á¶¡: 2014/5/15 to 2015/5/14
    • ÃöÁäµü:
    • ºK­n²¤¶:
      Loading effect is the process result deviation from layout difference. Different layout could have CD , topography , depth or other geometrical difference according to fab mass production experience. By current practice, routine and tedious manual check would be conducted for a new release product at every critical stage to avoid process deviation induced yield loss or device target offset. For example, line bridging due to etch CD bias difference from loading effect, or device target offset due to iso and dense area. With local layout information such as pattern density or line end density as an input to engineer, high risk area and corresponding process stages shall be identified in advance by machine-learning models proposed in this project.

  32. ¤j¶q´CÅé¯S¼x¸ê®Æ®w¤ñ¹ï·j´M§Þ³N

    • ­^¤å¦WºÙ: Matching and Retrieval Techniques for Large Multimedia Databases
    • ­pµe½s¸¹:
    • ¥D«ù¤H: ±i´¼¬P
    • ¸É§U³æ¦ì: ¸êµ¦·|
    • ­pµe°õ¦æ´Á¶¡: 2014/2/1 to 2014/12/15
    • ÃöÁäµü:
    • ºK­n²¤¶:
      ¥»­pµeÀÀ°ò©ó¥h¦~¡u¦Û°Ê¤º®e¿ëÃѧ޳N¡v¤§¦¨ªG¡A¨Ã¤Þ¶iGPUµ¥¥­¦æ§Þ³Nªº¤ä´©¡A¥H§Ö³t¦a¹ï¤j¶q´CÅé¸ê®Æ¶i¦æ³B²z¡C¬°¤F¹F¨ì¨Ï¥ÎªÌ¦b¬Ý¹qµø¸`¥Ø©Î¨ä¥L´CÅ餺®eªº¦P®É¡A´¼¼z¦æ°Ê¸Ë¸m¦P®É¥i¥H±Ò°Ê¬ÛÃöªºTV App¡A¬G¦Ó§Ú­Ì»Ý­n«Øºc¤@­Ó¤º®e¦Û°Ê¿ëÃѤÞÀº¡C¥»­pµe¥Ø¼Ð¬O¬ãµo¿ëÃѤÞÀº¤¤ªºÁn¯¾¯S¼x­ÈµÑ¨úºtºâªk¥H¤ÎÁn¯¾¤ñ¹ïºtºâªk¡A¨Ã¥HGPUµ¥¥­¦æ§Þ³N¶i¦æ¥[³t¤ñ¹ï¡A¥H³B²zÀH¿ïµø°Tµ¥ªA°È¤¤ªº¤j¶q¸`¥Ø¸ê®Æ¡C

  33. ´¹¶ê¯Ê³´¼Ë¦¡¿ë»{¤Î¬Û¦ü«×¤ÀªR

    • ­^¤å¦WºÙ: ¢åafer Failure Pattern Fingerprint and Similarity Detection
    • ­pµe½s¸¹:
    • ¥D«ù¤H: ±i´¼¬P
    • ¸É§U³æ¦ì: ¥x¿n¹q
    • ­pµe°õ¦æ´Á¶¡: 2013/9/1 to 2014/8/31
    • ÃöÁäµü:
    • ºK­n²¤¶:
      The objectives and scope of the project are:
      1. Develop and integrate a robust similar wafer detection methodology into wafer fingerprint kernel for CP/WAT failure diagnosis infrastructure.
      2. Enhance the stability, consistency and aaccuracy of failure pattern recognition system.
      3. Architecture design and prototyping for automation.

  34. ¥HGPU¬°¹Bºâ®Ö¤ß¤§­µ¼ÖÀ˯Á¨t²Î

    • ­^¤å¦WºÙ: GPU-based Music Information Retrieval Systems
    • ­pµe½s¸¹:
    • ¥D«ù¤H: ±i´¼¬P
    • ¸É§U³æ¦ì: °ê¬ì·|
    • ­pµe°õ¦æ´Á¶¡: 2013/8/1 to 2015/7/31
    • ÃöÁäµü:
    • ºK­n²¤¶:
      GPU¦b¦UºØ¬ì¾Ç­pºâªº­«­n©Ê¡A¤w¸g¤£¨¥¥i³ë¡Aªñ´Á³ÌÅãµÛªº½d¨Ò¡A¬O¤j³°¤Ñªe¤@¸¹Aµ²¦X¤F7,168ÁûNVIDIA Tesla? M2050 GPU©M14,336ÁûCPU¡A³Ð¤U°ª¹F2.507 petaflopsªº¥þ·s®Ä¯à¬ö¿ý¡A¦b2010¦~6¤ëTop500¶W¯Å¹q¸£±Æ¦W®³¨ì¥þ¥@¬É²Ä¤@¦W¡C¥»­pµe±N¥HGPU¬°¥D­n¹Bºâ®Ö¤ß¡A¨Ó¹ê§@¨âºØ­µ¼ÖÀ˯Áªº¨å«¬¡A¥]§t¡u­ó°Û¿ïºq¡v¡]query by singing/humming¡^©M¡u­µ°T«ü¯¾¡v¡]audio fingerprinting¡^¡A¨Ã±´°Q¦b¤£¦Pªº¤ñ¹ïµ¦²¤¤U¡A¦p¦ó¨ÏGPU+CPUªº¬[ºc¹F¨ì³Ì¦nªº®Ä¯à¡C¦P®É§Ú­Ì±N¨Ï¥Î¾÷¾¹¾Ç²ßªº¤èªk¡A´M·sªº¤ñ¹ï¤èªk¡A¦P®É¦b¤j¶q¸ê®Æ¤¤¾Ç²ß³Ì¦nªº¤ñ¹ïµ¦²¤¡C§Ú­Ì´Á±æ¯à°÷¨Ï¥Î§C·GªºµwÅé¤Î¸û§Cªº¯Ó¹q¶q¡A´N¯à¶i¦æ¤j¶q­µ¼Öªº¤ñ¹ï¡]¡u­ó°Û¤ñ¹ï¡v¯à°÷¦b5¬í¤º¤ñ¹ï5¸U­ººq¡A¡uÁn­µ«ü¯¾¡v¯à°÷¦b¤­¬í¤º¤ñ¹ï20¸U­ººq¡^¡A¨Ã¯à°÷¦³²Å¦X¥@¬É¤ô·Çªº¿ëÃѲv¡]¥HMIREX¤ñÁɬ°¤ñ¸û°ò·Ç¡^¡A¥H±À¶i°ê¤º¬ÛÃö§Þ³N»PÀ³¥Îªº¤ô·Ç¡C

  35. ¦Û°Ê¤º®e¿ëÃѧ޳N

    • ­^¤å¦WºÙ: Automatic Content Recognition Technology
    • ­pµe½s¸¹:
    • ¥D«ù¤H: ±i´¼¬P
    • ¸É§U³æ¦ì: ¸êµ¦·|
    • ­pµe°õ¦æ´Á¶¡: 2013/2/1 to 2013/12/15
    • ÃöÁäµü:
    • ºK­n²¤¶:
      ÀHµÛ¤â«ù´¼¼z¸Ë¸mªº´¶¤Î¡A«Ü¦hÆ[²³¬Ý¹qµø®É¡A³£·|¦P®É¨Ï¥Î´¼¼z¦æ°Ê¸Ë¸m¥H¾É¤J¬ÛÃö¸ê°T¡A«P¦¨¬ÛÃöApp(´¼¼z¦æ°Ê¸Ë¸m¤W¤§À³¥Îµ{¦¡)ªº¿³°_¡C¥Ø«eªºApp¤j¦h»Ý­nÆ[²³¦Û¦æ¶}?¸Óµ{¦¡¡A¥BÆ[²³Âà¥x®É¡AApp ¨Ã¤£·|¸òµÛ¦Û°ÊÂà¥x¡C¦Ó¬°¹F¨ìÅý¨Ï¥ÎªÌ¦b¬Ý¹qµø¸`¥Øªº¦P®É¡AApp ¯à¸òµÛ¦Û°ÊÂà¥x¡A¬Æ¦Ü´¼¼z¦æ°Ê¸Ë¸m(i.e. Smart Phone, xPad)¥»¨­¦P®É¥i¥H±Ò°Ê¬ÛÃöªºTV App¡A©Ò¥H§Ú­Ì»Ý­n«Øºc¤@­Ó¹qµø¤º®e¦Û°Ê¿ëÃѤÞÀº¡C¦]¦¹¡A¥»­pµe¥Ø¼Ð¬O¬ãµo¿ëÃѤÞÀº¤¤ªºÁn¯¾¯S¼x­ÈµÑ¨úºtºâªk¡A¥H¤ÎÁn¯¾¤ñ¹ïºtºâªk¡C

  36. ±q¼Æ¦ì¾Ç²ß¨ì´¼¼z¥Í¬¡ªº¾ã¦X¬ãµo­pµe

    • ­^¤å¦WºÙ: Integrated R&D Program from Digital Learning to Smart Life
    • ­pµe½s¸¹:
    • ¥D«ù¤H: ±i´¼¬P
    • ¸É§U³æ¦ì: °ê¬ì·|
    • ­pµe°õ¦æ´Á¶¡: 2013/1/1 to 2015/11/30
    • ÃöÁäµü:
    • ºK­n²¤¶:
      ¥»­pµe¹Á¸Õ¾ã¦X¦U­pµe¥D«ù¤Hªº±Mªø¡A²Ö¿n¹L¥hªº¶}µo¯à¶q¡Aµ²¦X¼t°Ó¹ê»Ú»Ý¨D¡A¶}µo·s¤@¥Nªº»y­µ§U²z¾ã¦X¤¶­±¡C¾ã¦X¥x¤j±i´¼¬P±Ð±Âªø´Á¦b»y­µ¤è­±ªº±Mªø¡A³z¹L¨ä¤£¦P»y¨¥ªº»y­µ¿ëÃÑÃöÁäµüÂ^¨ú(Keyword-Spotting)¡A¥H¤Î±Û«ß¿ëÃÑ¡Aµ²¦X³\»D·G¯S¸u¬ã¨s­ûªø´Á¦b¦ÛµM»y¨¥²z¸Ñªº²`¤J¬ã¨s»P¸êµ¦·|¦b¤£¦P¸Ë¸m¶¡ªº­Ó¤H¤Æ¾Ç²ß§Þ³N¡A¾ã¦X¶}µo·s¤@¥Nªº»y­µ§U²z¡A¥H´Á¬°¥xÆWªº·~ªÌ(¥Ø«e¬¢½Í¤¤¥]¬AÂE®ü¬ì§Þ¡A»·¨£¬ì§Þ¡AÁÉ·L¬ì§Þ/Cyberon¡A...µ¥)½Ñ¦h¼t°Ó¡A´£¨Ñ¤@­Ó§¹¾ãªº¾ã¦X»y­µ§U²z¥­¥x¡C

  37. ´¹¶ê¯Ê³´¼Ë¦¡¿ë»{

    • ­^¤å¦WºÙ: Wafer Failure Pattern Fingerprint
    • ­pµe½s¸¹:
    • ¥D«ù¤H: ±i´¼¬P
    • ¸É§U³æ¦ì: ¥x¿n¹q
    • ­pµe°õ¦æ´Á¶¡: 2013/1/1 to 2013/12/31
    • ÃöÁäµü:
    • ºK­n²¤¶:
      Failure analysis is the process of analyzing and interpreting wafer data to identify the root cause of a failure. This module plays a pivotal role in enhancing the wafer yield. To ensure that the failure analyzer delivers meaningful results, a robust wafer failure pattern recognition platform with efficient similarity wafer ranking should be available. Wafer failure pattern recognition, a first step in the direction, is meant to detect and recognize the appropriate failure lots/wafers which can be used for failure correlation during the analysis stage. Wafer pattern similarity is the next milestone in the path leading to failure analysis. Failure pattern Similarity ranking assists in performing failure analysis with a higher degree of consistency. A logical inference with respect to suspected tools is possible only when the failure pattern recognition is equipped with pattern similarity ranking. In this proposal, we plan to build up a computer aided tool to simplify and speed up the pattern similarity ranking and recognition instead of manual check CP/WAT failure map by eyeball view, and enhance both high engineering efficiency and effectiveness.

  38. ¤f»¡¥x»yµû¤À¨t²Î¤§¬ã¨s»P¹ê§@

    • ­^¤å¦WºÙ: Research and Implementation of Spoken Taiwanese Scoring System
    • ­pµe½s¸¹:
    • ¥D«ù¤H: ±i´¼¬P
    • ¸É§U³æ¦ì: °ê¬ì·|
    • ­pµe°õ¦æ´Á¶¡: 2012/8/1 to 2013/7/31
    • ÃöÁäµü:
    • ºK­n²¤¶:
      ¥»­pµeªº¬ãµo¥Ø¼Ð¡A¬O­n§¹¦¨¤@­Ó§¹¾ãªº¥x»yCAPT¡]¹q¸£»²§Uµo­µ°V½m¡Acomputer assisted pronunciation training¡^¨t²Î¡A©Ò¨Ï¥Îªºµû¤À°Ñ¼Æ¥]§t­µ¦â¡B­µ½Õ/­µ°ª¡B­µ¶q/¯à¶q¡B­µªø/Ãý«ßµ¥¡A¦P®É§Ú­Ì¤]±N±´°Q¬ÛÃöªº¬ã¨sijÃD¡A¨Ò¦p¦p¦ó¦Û°Ê¶i¦æ²V²c­µªº°»´ú¡B¦p¦óµ²¦X±j¨î¹ï¦ì¡]forced alignment¡^¤Î¦Û¥Ñ­µ¸`¸Ñ½X¡]free syllable decoding¡^¨Ó±o¨ìí©wªºµû¤À¤À¼Æ¡B¦p¦ó¶i¦æ¥x»yÁn½Õ¿ëÃÑ¡A¥H¤Î¦p¦ó½Õ¾ã³o¨Çµû¤À°Ñ¼ÆªºÅv­«¡A¥H«K¯à°÷¹Gªñ¦Ñ®v©Òµ¹ªº¹ê»Ú¤À¼Æµ¥µ¥¡C¦¹¨t²Î¥²¶·¯à°÷­pºâ¨C¤@­Óµü·J¤Î¨C¤@­Ó­µ¯Àªº¤À¼Æ¡A¨ÃÅã¥Ü¬ÛÃöªº²V²c­µ¡A¦P®Éµ¹¤©µo­µ§ï¶iªº«Øij¡AÅý¨Ï¥ÎªÌ¯à°÷¤ÏÂнm²ß¡A¥[±j¦Û¤v¤f»¡¥x»yªº¥¿½Tµo­µ¡C¡u¹q¸£»²§Uµo­µ°V½m»Pµû¤À¡v¬O»y­µ¿ëÃѪº¤@­Ó·s¿³¬ã¨s»PÀ³¥Î»â°ì¡A¬ÛÃöªº¤åÄm»P³ø§i¤éº¥Â×´I¡A³nÅéÀ³¥Î¤]¶V¨Ó¶V¦h¡A¦ý¤´¥¼¨£»P¥x»y¬ÛÃöªºÀ³¥Î¡C§Ú­Ì¦b¥ý«eªº²£·~¦X§@­pµe¤¤¡A¤w¸g³°Äò§¹¦¨¤F¡uµØ»y»y­µµû¤À¡v¡B¡u­^»y»y­µµû¤À¡v¡B¡u¤é»y»y­µµû¤À¡vµ¥¨t²Î¡A¦Ó¥B¤]¶i¦æ¬ÛÃöªº§Þ³NÂಾ¤Î°Ó«~¤Æ¡A¬ãµo¤Î²£·~¦X§@ªº¸gÅç¬Û·íÂ×´I¡C¦b¦¹­pµe¤¤¡A§Ú­Ì§Æ±æµ²¦X¨Ã¤Þ¥Î¨ä¥L¤l­pµe¥D«ù¤H¦b¥x»y»y¨¥³B²zªº¦h¦~¬ã¨s¦¨ªG¡A¹ê»Ú²£¥X¤@­Ó±µªñ¥«³õ»Ý¨Dªº¡u¹q¸£»²§U¥x»yµo­µµû¤À¨t²Î¡v¡A¦]¦¹¦¹­pµeªº¨Ï©R°£¤F¦b©ó¯à°÷¥H»y­µ¬ì§Þ¨Ó«P¶i¼Æ¦ì¾Ç²ß²£·~¤É¯Å¥~¡A§ó¨ã¦³¥»¤g¤å¤Æ±À¼sªº²`¤@¼h·N¸q¡C

  39. ±m§©Âಾªº¹ê§@»P±´°Q

    • ­^¤å¦WºÙ: The Practice and Implementation of Makeup Transfer
    • ­pµe½s¸¹:
    • ¥D«ù¤H: ±i´¼¬P
    • ¸É§U³æ¦ì: °ê¬ì·|¡B³Ð·NÅÚ½³
    • ­pµe°õ¦æ´Á¶¡: 2012/6/1 to 2013/5/31
    • ÃöÁäµü:
    • ºK­n²¤¶:
      ¥»³ø§i¤º®e²­z¦¹­pµe¹ï©ó¦Û°Ê±m§©ªºÀ³¥Î¡A¥Ø«eªºµo®i¶i«×»P¦¨ªG¡C¥»­pµe»EµJ¦bÀ³¥Î©ó´¼¼z«¬¥­¥x¤Wªº¦Û°Ê²´½u°lÂÜ¡A¦]¬°²´½u¬O­«­nªº¤HÁy¯S¼x¡A¦b¤Æ§©®É­Y¦³µe¤W²´½u©¹©¹·|¦³µeÀsÂI·úªº®ÄªG¡A©Ò¥H¥»­pµe±NÀu¥ý±´°Q²´½u°lÂܧ޳N¡C¦b¬ã¨s¤èªk¤W«h¤À¬°¤G­Ó³¡¤À¡A¤@¬O¤HÁy°»´ú»P²´·ú°»´ú¡A¥t¤@«h¬O²´½u°lÂÜ¡C¦b¸ê®Æ¦¬¶°¤W¡A§Ú­Ìªº¸ê®Æ®w¦¬¶°¤F179±iÁy³¡·Ó¤ù¡A¨Ã¤H¤u¼Ð°O²´½u¦ì¸m¡C¹êÅçµ²ªGÅã¥Ü¡A²´½u°lÂܪº¦¨¥\²v¶W¹L90%¡C³Ì«á§Ú­Ì¤]¦¨¥\¶}µo¤F¤@­Ó¥i¥H¥Î©óAndroid¥­¥xªºAPP¡AÃÒ¹ê¤F¦Û°Ê²´½u°lÂÜÀ³¥Î¦b´¼¼z«¬¥­¥xªº¥i¦æ©Ê¡C

  40. ­µ¼Ö¶i¶¥¯S¼x©â¨ú»P¤H¾÷¤¬°Ê§Þ³N

    • ­^¤å¦WºÙ: Technologies for Advanced Music Feature Extraction and Human¡VComputer Interaction
    • ­pµe½s¸¹:
    • ¥D«ù¤H: ±i´¼¬P
    • ¸É§U³æ¦ì: ¤¤µØ¹q«H¬ã¨s©Ò
    • ­pµe°õ¦æ´Á¶¡: 2012/1/1 to 2012/12/31
    • ÃöÁäµü: ­ó°Û¿ïºq¡BºqÁnµû¤À¡B¦±­·¤ÀÃþ¡B¸`©ç°lÂÜ¡B­µ°T­µ¼Öªº­µ°ª°lÂÜ
    • ºK­n²¤¶:
      ¥»­p¹º¦®¦b¾ã¦X­µ¼Ö¶i¶¥¯S¼xªº©â¨ú¡A¨Ã»Pºq°Û¿ëÃѧ޳N»Pºq°Ûµû¤À§Þ³N°µµ²¦X¡A¹ê²{°ò©ó¦¹©Ò²£¥Íªº¤H¾÷¤¬°Ê¨t²Î¡C¦bºq°Û¿ëÃѨt²Î¤è­±¡A­º¥ý¥²¶·¶i¦æ¶i¶¥­µ¼Ö¯S¼xªº©â¨ú¥H«Ø¥ß­µ¼Ö¸ê®Æ®w¡A ¹ï©ó±ý«Ø¥ß¸ê®Æ®wªºMP3/MIDI­µ¼ÖÀÉ¡A±Ä¥Î¡u°»´úºqÁn¤ù¬q¡v§Þ³N¼Ðµùºq¦±¤¤¦³¤HÁnªº¤ù¬q¡A¶i¥H¨Ï¥Î¡uºq°Û¥D±Û«ß©â¨ú§Þ³N¡v¨Ó¤Á³Î²VÂøµÛ­µ¼Ö»P¤HÁnªº¤ù¬q¡A¨Ã±N¨ä¤¤ªº¤HÁn±Û«ß©â¨ú¥X¨Ó¡A¥H´£¨Ñ«áÄòªººq°Û¿ëÃѨt²Î¨Ï¥Î¡F¦b«eºÝ³¡¥÷¡A¹ï©ó¿é¤Jªº¤HÁn­µÀÉ¡A±Ä¥Î¡u±j°·©Êºq°Û¿ëÃѤñ¹ï§Þ³N¡v¡A¦¹§Þ³N¥i±N¿é¤Jªº¤HÁn­µÀÉ»P­µ¼Ö¸ê®Æ®w°µ¤ñ¹ï¡A¨Ã¦^¶Ç¤ñ¹ïµ²ªG°µ¶i¤@¨BªºÀ³¥Î¡C¹ï©ó¡uºq°Ûµû¤À§Þ³N¡v¦Ó¨¥¡A±N¨Ï¥Î«e­z¬ÛÃö­µ¼Ö¶i¶¥¯S¼x§Þ³N¡A¹F¦¨©â¨ú­µ¼ÖMV¤§¥D±Û«ßªº¥Øªº¡A¨Ã¹ï¨Ï¥ÎªÌ©Ò¿é¤J¤§ºq°Û°T¸¹¶i¦æµû¤Àªº¤u§@¡A¼W¥[¤H¾÷¤¬°Êªºµo®i¡C

  41. ³z¹L»y­µ»PÃöÁä¦r²Õªº¹qµø¸`¥Ø¦Û°Ê¸ê®ÆµÑ¨ú¤èªk

    • ­^¤å¦WºÙ: Method for Automatic Data Extraction of TV Programs by Voice and Keyword Groups
    • ­pµe½s¸¹:
    • ¥D«ù¤H: ±i´¼¬P
    • ¸É§U³æ¦ì: ¸êµ¦·|
    • ­pµe°õ¦æ´Á¶¡: 2012/1/1 to 2012/12/15
    • ÃöÁäµü:
    • ºK­n²¤¶:
      ©Ò¿×»y­µ¤å¥óÀ˯Á¡A¬O¯à±µ¨ü¨Ï¥ÎªÌªº»y­µ¬d¸ß¡A¨Ó¹ï¸ê®Æ®w¤¤ªº¤å¥ó¶i¦æÀ˯Á¡C©ó»y­µ¿ëÃѪº³¡¤À¡A±`¥Îªº¤èªk¦³ÃöÁäµüµÑ¨ú¡A¥H¤Î¦Û¥Ñ­µ¸`¸Ñ½Xµ¥µ¥¡C¦Ó¦bÀ˯Á³¡¤À¡A¹ï©ó¸ê®Æ®w¬O¤å¦r©Î»y­µªº«¬ºA¡A¤]¦³¤£¦Pªº¤èªk¡G­Y¬O¤å¦r«¬ºA¡A«h³z¹LÂ_µüªº¤è¦¡¡A±N¸g±`¥X²{ªºµü·J©w¸q¬°ÃöÁä¦r¡F¦Ó­Y¬O»y­µ«¬ºA¡A«h­n¥ýÂà´«¬°¤å¦r¡A¥H¥[³t¨Ï¥Î®Éªº³t«×¡CÃö©óÃöÁäµüµÑ¨ú¡A­º¥ý¬O¿ëÃÑ»y¥y·í¤¤¡A¬O§_¦s¦b¯S©wªºÃöÁä¦r¡F­Y¦³¡A«h±N»y¥y¤¤¥]§tÃöÁäµüªº³¡¤ÀµÑ¨ú¥X¨Ó¡C¤@¯ë·|¦b¯S©w¥ô°È¡]Task¡^¤U¡A¹ï¥ô°È¿ï¨ú­Y¤z­ÓÃöÁäµü¡A¦b¿ëÃѮɥu­n±NÃöÁäµüµÑ¨ú¥X¨Ó¡A¦Ó¤£ºÞ¨ä¥¦³¡¤À¡C¦]À³¥ô°Èªº¤£¦P¡AÃöÁäµüªº©w¸q¤]´N¤£¦P¡A¦Ó©Ò·f°tªºµLÃöµü¼Ò«¬¤]«Ü¥i¯à´N¤£¦P¡C¤@­Ó¦nªºÃöÁäµüµÑ¨ú¨t²Î¥²¶·¹F¨ì¤H©Ê¤Æªº­n¨D¡AÅý¨Ï¥ÎªÌ¥ô·N¦a¹B¥Î¦b¬Y¨Ç¥ô°È¤¤¡A»¡¥X¨Ï¥ÎªÌ·Q­nÁ¿ªº»y¥y¡A¤£¯à¥[¥H­­¨î¡CÃöÁäµüµÑ¨ú¨t²Î¤¤ªº¿ëÃѺô¸ô°ò¥»¤W¬O¥ÑÃöÁäµüºô¸ô©MµLÃöµüºô¸ô¨â¤j³¡¤À©Ò²Õ¦X¦Ó¦¨¡A§Ú­Ì¥i¥H¨Ì¾Ú±¡¹Òªº»Ý¨D¡A¨Æ¥ý©w¸q¦nÃöÁäµü·J¡A¨Ã»PµLÃöµüªº³¡¤À°µ«ê·íªº²Õ¦X¡C¦Ó¦b¿ëÃѮɡA³q±`¥i±Ä¥Î³sÄò»y­µ¿ëÃѧ޳N¨Ó¶i¦æÁn¾Ç¼h¦¸ªº¿ëÃÑ¡C³Ì±`±Ä¥Îªº§Þ³N¬Oºû¯S¤ñ·j´Mªk¡A¯à°÷¦P®É¹ï»y­µ«H¸¹§@­µ¸`ªº¤Á³Î»P¿ëÃÑ¡A¬Û·í¾A¦XÀ³¥Î¦b§Y®É¨t²Î¡C

  42. ´¹¶ê¯Ê³´¹Ï¹³¤§¤ÀªR»P¿ëÃÑ

    • ­^¤å¦WºÙ: Wafer Failure Pattern Recognition
    • ­pµe½s¸¹:
    • ¥D«ù¤H: ±i´¼¬P
    • ¸É§U³æ¦ì: ¥x¿n¹q
    • ­pµe°õ¦æ´Á¶¡: 2011/9/1 to 2012/8/31
    • ÃöÁäµü: ´¹¶ê¯Ê³´¤ÀªR¡B¼Ë¦¡¿ë»{¡B¾÷¾¹¾Ç²ß¡B¼v¹³³B²z¡B¾Ç²ß±Æ§Ç
    • ºK­n²¤¶:
      ´¹¶ê¦b¸g¹L¦U¦¡¦U¼Ëªº»sµ{«á¡A·|¥Ñ©ó¾÷¥xªºª¬ªp¦Ó²£¥Í¤£¦Pªº¯Ê³´¹Ï§Î¡A¥»­pµe¹Á¸Õ¨Ï¥Î¦UºØ¤£¦Pªº¤èªk¡]¥]§t¼v¹³³B²z¡B¾÷¾¹¾Ç²ß¡B¼Ë¦¡¿ë»{¡B¾Ç²ß±Æ§Çµ¥¡^¨Ó¤ÀªR´¹¶êªº¯Ê³´¹Ï¹³¡A¨Ã¹ï´¹¶ê¯Ê³´¹Ï¹³¶i¦æ¤ÀÃþ¡A¥ø¹Ï¶i¤@¨B§ä¥X¾É­P¯Ê³´ªº¾÷¥x¬G»Ù­ì¦]¡C

      Failure analysis is the process of collecting and analyzing data to determine the cause of a failure. It is an important discipline in many branches of manufacturing industry. Wafer failure pattern recognition is the first data mining step to detect the failure lots/wafers for failure correlation. Pattern recognition is generally categorized according to the type of learning procedure used to generate the recognition result. In this proposal, we will build up one computer aided tool to simplify and speed up the pattern recognition instead of manual check CP/WAT failure map by eyeball view, and enhance both high engineering efficiency and effectiveness.

  43. ¤À´²¦¡¶³ºÝ¤¤¤¶³nÅé¡]E2¤l­pµe¡G¶³ºÝÀ³¥Î§G¸p»PºÞ²z¨t²Î¡^

    • ­^¤å¦WºÙ: Cloud Application Deployment and Management System
    • ­pµe½s¸¹:
    • ¥D«ù¤H: ±i´¼¬P
    • ¸É§U³æ¦ì: ¥x¹F¹q
    • ­pµe°õ¦æ´Á¶¡: 2011/8/1 to 2014/7/31
    • ÃöÁäµü:
    • ºK­n²¤¶:

  44. °ò©ó¼v¹³¤§Áy³¡½§½è¤ÀªR»P·å²«­×¸É

    • ­^¤å¦WºÙ: Image-based facial skin analysis and flaws covering
    • ­pµe½s¸¹: NSC 100-2622-E-007 -007 -CC2
    • ¥D«ù¤H: ±i´¼¬P
    • ¸É§U³æ¦ì: °ê¬ì·|¡B³Ð·NÅÚ½³
    • ­pµe°õ¦æ´Á¶¡: 2011/6/1 to 2012/5/31
    • ÃöÁäµü: ½§½è¤ÀªR¡B·å²«­×¸É¡B§÷½è¤ÀªR¡B§÷½è¦X¦¨
    • ºK­n²¤¶:
      ²{¥N¤H´¶¹M´÷±æÅܱo§ó¬üÄR¡A¤]³ßÅw¥ÎÃèÀY¨Ó°O¿ý¬üªº¨Æª«¡C¥»­pµeªº¥Ø¼Ð¡A­n¶}µo¤@­Ó¾ã¦X©Êªº½§½è¤ÀªR»P·å²«­×¸É¨t²Î¡A¨Ã²¾´Ó¨ì´O¤J¦¡¥­¥x¤W(¨Ò¦p´¼¼z«¬¤â¾÷)¡A¹ï©ó¨S¦³¤Æ§©ªº¨Ï¥ÎªÌ¦Ó¨¥¡A¥i¥H¤è«KªººÊ±±»PºÞ²z¦Û¤vÁy³¡¦Ù½§ªº°·±d¡A¹ï©ó¦³¤Æ§©ªº¨Ï¥ÎªÌ¦Ó¨¥¡A¥i¥H§Ö³tÀË´ú§©¬O§_µo¥Í¤£§¡¤Ãªº±¡ªp¡C¹ï©ó³ßÅw¦Û©çªº¨Ï¥ÎªÌ¡A¥i¥H¦Û°Ê¹ïÁy¤Wªº·å²«¶i¦æ­×¸É¡C¦]¦¹¡A§Ú­Ì±N¥Î§÷½è¤ÀªRªº¤èªk¨Ó¿ëÃÑÁy¤Wªº¦UºØ·å²«¯S¼x(¥]§t½K¯¾¡Bµkµk¡B¯»¨ë)¡A¥H¤Î¤ÀªR¦Ù½§ªº¥ú·Æ«×¡A³Ì«á¦A±N·å²«°Ï°ì¥Î§÷½è¦X¦¨ªº¤èªk¶i¦æ­×¸É¡CµM¦Ó§{¶¡³\¦h¬Û¾÷¤w¤º«Ø¬ü½§¼Ò¦¡¡AÁöµM¥i¥H­×±¼³¡¤À·å²«¡A¦ý¬O³\¦h«D·å²«°Ï°ìªº²Ó¸`¥i¯à¤]·|³Q¼Ò½k±¼¡A¦Ó¥»­pµe¥u°w¹ï·å²«°Ï°ì¶i¦æ­×¸Éªº¤èªk¬Û«H±N¥i¹F¨ì§ó¦nªº®ÄªG¡C§Ú­Ì§Æ±æ¸g¥Ñ¥»­pµe¡A¹ê»Ú¶}µo¥X¤@­Ó¾A¥Î©ó´O¤J¦¡¥­¥xªº¿ëÃѮ֤ߡA¦A±À¼s¨ì·~¬É¡A¥Ñ©ó´¼¼z«¬¤â¾÷¬O¥¼¨ÓªºÁͶաA§Ú­Ì´Á³\¯à¬°¨Ï¥ÎªÌ±a¨Ó§ó¤è«KªºªA°È¡C

  45. ¤ä´©Cloud-aware´O¤J¦¡¦æ°Ê¦h®Ö¤ß¥­¥x--¤l­pµe¤T¡G¾ã¦X´O¤J¦¡¨t²Î»P¶³ºÝ­pºâªº­µ¼Ö»P»y­µªA°È

    • ­^¤å¦WºÙ: Supporting voice/music services for mobile & cloud synergism
    • ­pµe½s¸¹: NSC 100-2219-E-007 -008
    • ¥D«ù¤H: ±i´¼¬P
    • ¸É§U³æ¦ì: °ê¬ì·|
    • ­pµe°õ¦æ´Á¶¡: 2011/5/1 to 2013/4/30
    • ÃöÁäµü: »y­µ¿ëÃÑ¡B­ó°ÛÀ˯Á¡BºVÀ»À˯Á¡B¦±­·¤ÀÃþ¡B¶³ºÝ¹Bºâ¡B¥­¦æ³B²z
    • ºK­n²¤¶:

      ÀHµÛµL½uºô¸ô¥H¤Î¦æ°Ê¸Ë¸m¼½©ñ­µ¼Ö¥\¯àªº´¶¤Î¡A¨Ï­µ¼Ö»P»y­µªA°È¦¨¬° ¦æ°Ê¥[­Èªº°ò¥»À³¥Î¤§¤@¡A¦Ó´£¨ÑªA°Èªº§Î¦¡¥ç¤éÁͦh¤¸¡A¦b»y­µ¤è­±¡A¤£ºÞ ¬O§Q¥Î¦æ°Ê¸Ë¸m·j´Mºô¸ô¤WªºÃöÁä¦r©Î¬O½s¼g¤â¾÷²°T¡A¤H­Ì§Æ±æ¥i¥H§Q¥Î»y ­µ¿é¤J¨ú¥N¦æ°Ê¸Ë¸mªº«öÁä¿é¤J¡A¹ï©ó³o¶µÀ³¥Î¡A»y­µ¿ëÃѪº³t«×¤Î·Ç½T©Ê«h ¦ÜÃö­«­n¡F¦b­µ¼Ö¤è­±¡A»P¹êÅé­µ¼ÖCD¬Û¸û¡A¤H­Ì§ó§Æ±æ¯àÀH®ÉÀH¦a§ä¨ì³Ì·Q Å¥¨ìªº­µ¼Ö¡C¦]¦¹¡A§Ú­Ì§Æ±æ«Øºc¤@­Ó¦³®Ä²vªº¨t²Î¡A¥i¥H¦b¦æ°Ê¸Ë¸m¤W³z¹L ¦UºØÀ˯Á¤è¦¡·j´M¤W¸U­ºªººq¦±¸ê®Æ®w¡A§ä¨ì¨Ï¥ÎªÌ·QÅ¥ªº­µ¼Ö¡A¦A¥[¤W¦p¸` «µ¡B¦±­·µ¥¯S¼xªº»²§U¡A±ÀÂ˨ϥΪ̥i¯à·|³ßÅwªº¦±¥Ø¡A¥Hº¡¨¬¨Ï¥ÎªÌ­Ó¤H³ß ¦n¡C

      ¥»­pµeÀÀ§Q¥Î¦æ°Ê¦h®Ö¤ß¥­¥xµo®i­µ¼Ö»P»y­µ¬ÛÃö¤§À³¥ÎªA°È¡A¨Ï¥ÎªÌ¥i ³z¹L¦UºØ¤è¦¡·j´M­µ¼Ö¡A¦pÂǥѭó°Û­µ¼Öªº¤ù¬q¡B»y­µ¿é¤Jºq¦W¡B±¡ºü©ÎºVÀ» ¸`«µ¶i¦æ·j´M¡C­º¥ý¥Ñ¤â«ù¦¡¸Ë¸m¹ï¿é¤Jªº­µ°T¶i¦æ«eºÝ³B²z¥H¨ú±o­µ°T¯S ¼x¡A¤§«á§Q¥Î¶³ºÝ§Þ³N¦b¦øªAºÝ¶i¦æ¸ê®Æ®w¤ñ¹ï¡A¦P®É·j´M¬Ûªñ¦±­·ªººq¦±±À Â˵¹¨Ï¥ÎªÌ¡A³Ì«á±N©Ò¦³¸ê°T¦^°e¦Ü¥Î¤áºÝ¨Ã§e²{µ¹¨Ï¥ÎªÌ¡C¦¹¬[ºcÁקK¤Fª½ ±µ¦b«eºÝ¸Ë¸m¤W¶i¦æ¤ñ¹ïªºÃe¤j¹Bºâ¶q¥H¤Î¸ê®Æ®wªºÀx¦sªÅ¶¡¡A¦]¦¹¥i¥H¾A¥Î ©ó¤j³¡¤Àªº¤â«ù¦¡¸Ë¸m¡A¦¹¥~¡A¦³Å²©ó©â¨ú¬Y¨Ç­µ°T¯S¼xªº¹Bºâ¶q¹ï©ó¤â«ù¦¡ ¸Ë¸mªº­t¾á¤´¤Ó­«¡A¦]¦¹§Q¥Î¦h®Ö¤ß¨t²Î¶i¦æ¥­¦æ³B²z¡A¹w´Á¥i¤j´T§ïµ½­pºâ ©Ò»Ýªº®É¶¡¡A¥[³t¾ãÅé³B²zªº³t«×¡A¼W¶iÀ³¥ÎªA°Èªº»ù­È¡C¬ã¨sªº¤u§@¶µ¥Ø¦p ¤U¡G

      1. «Ø¥ß¦h­«¯S¼x¿ëÃѪº­µ¼Ö·j´M¨t²Î
      2. »`¶°°µ¬°¿ëÃѥγ~ªº­µ¼Ö¤Î»y®Æ
      3. ¹ê§@­Ó¤H¤Æ­µ¼Ö±ÀÂ˨t²Î
      4. ¾ã¦X¨t²Î¨Ã¹ê§@©ó¦h®Ö¤ß´O¤J¦¡¥­¥x
      5. ´ú¸Õ¤Îµû¦ôÀ³¥ÎªA°È©ó¶³ºÝ¹Bºâªº®Ä¯à

  46. ®É¶¡§Ç¦C¦æ¬°±´°É§Þ³N

    • ­^¤å¦WºÙ: Temporal and Sequential Activity Mining
    • ­pµe½s¸¹:
    • ¥D«ù¤H: ±i´¼¬P
    • ¸É§U³æ¦ì: ¸êµ¦·|
    • ­pµe°õ¦æ´Á¶¡: 2011/1/1 to 2011/12/31
    • ÃöÁäµü: ¨Ï¥ÎªÌ·N¹Ï¤ÀªR¡B¸sÅé¦æ¬°¹w´ú¡B§Ç¦C¦æ¬°¤ÀªR
    • ºK­n²¤¶:
      ¥»¦X§@¬ã¨s­pµe¹w­pµo®i¥X®É¶¡§Ç¦C¦æ¬°±´°É©M¹w´ú¼Ò²Õ¡A¥Î¥H»²§U¬ì±M­pµe¡uInteractive Consumer Intention Analysis Engine¡v¤ÀªR¨Ã¹w´ú¨Ï¥ÎªÌ·N¹Ï(©Î¦æ¬°)¡C¦b¤ÀªR¨Ï¥ÎªÌ·N¹Ï©Î¦æ¬°¤è­±¡A§Ú­Ì·|¥ý»`¶°(1)¸sÅé¦æ¬°¸ê®Æ(¦p³¡¸¨«È¦bªÀ¸sºô¯¸¤À¨Éªº¤ß±o)¤Î(2)­Ó¤H¦æ¬°¸ê®Æ(¦p®ø¶OªÌ­Ó¤H°¾¦n)¡A¨Ã§Q¥Î¸ê®Æ±´°É§Þ³N(¦pNºû®É¶¡§Ç¦C¦æ¬°±´°É§Þ³N)¤ÀªR¥X¸sÅé/­Ó¤H¦æ¬°¼Ò¦¡¡A¨Ã¶}µo¥X®É¶¡§Ç¦C¦æ¬°¹w´ú¼Ò²Õ¡A¥H»²§U¥»Åé­pµe§@¬°¥¼¨Ó¹w´ú¨Ï¥ÎªÌ·N¹Ï©Î¦æ¬°¤§À³¥Î¡C

  47. ¥x»y»y­µ»P¤å¦r¦h­±¦V»y®Æ®w¤§«Ø¸m¤Î¨ä¦b¥x»y­pºâ»y¨¥¾Ç¤§À³¥Î--¤f»¡¥x»yµû¤À¨t²Î¤§¬ã¨s»P¹ê§@

    • ­^¤å¦WºÙ: Corpus Collection for Taiwanese Texts and Speech with Applications to Taiwanese Computational Linguistics - The Research and Development of Spoken Taiwanese Scoring Systems
    • ­pµe½s¸¹: 99-2221-E-007-049-MY3
    • ¥D«ù¤H: ±i´¼¬P
    • ¸É§U³æ¦ì: °ê¬ì·|
    • ­pµe°õ¦æ´Á¶¡: 2010/8/1 to 2013/7/31
    • ÃöÁäµü: ¹q¸£»²§Uµo­µ°V½m¡B¹q¸£»²§U»y¨¥¾Ç²ß¡B»y­µ¿ëÃÑ¡B»y­µµû¤À¡BÁn½Õ¿ëÃÑ
    • ºK­n²¤¶:
      ¥»­pµeªº¬ãµo¥Ø¼Ð¡A¬O­n§¹¦¨¤@­Ó§¹¾ãªº¥x»yCAPT¡]¹q¸£»²§Uµo­µ°V½m¡Acomputer assisted pronunciation training¡^¨t²Î¡A©Ò¨Ï¥Îªºµû¤À°Ñ¼Æ¥]§t­µ¦â¡B­µ½Õ/­µ°ª¡B­µ¶q/¯à¶q¡B­µªø/Ãý«ßµ¥¡A¦P®É§Ú­Ì¤]±N±´°Q¬ÛÃöªº¬ã¨sijÃD¡A¨Ò¦p¦p¦ó¦Û°Ê¶i¦æ²V²c­µªº°»´ú¡B¦p¦óµ²¦X±j¨î¹ï¦ì¡]forced alignment¡^¤Î¦Û¥Ñ­µ¸`¸Ñ½X¡]free syllable decoding¡^¨Ó±o¨ìí©wªºµû¤À¤À¼Æ¡B¦p¦ó¶i¦æ¥x»yÁn½Õ¿ëÃÑ¡A¥H¤Î¦p¦ó½Õ¾ã³o¨Çµû¤À°Ñ¼ÆªºÅv­«¡A¥H«K¯à°÷¹Gªñ¦Ñ®v©Òµ¹ªº¹ê»Ú¤À¼Æµ¥µ¥¡C¦¹¨t²Î¥²¶·¯à°÷­pºâ¨C¤@­Óµü·J¤Î¨C¤@­Ó­µ¯Àªº¤À¼Æ¡A¨ÃÅã¥Ü¬ÛÃöªº²V²c­µ¡A¦P®Éµ¹¤©µo­µ§ï¶iªº«Øij¡AÅý¨Ï¥ÎªÌ¯à°÷¤ÏÂнm²ß¡A¥[±j¦Û¤v¤f»¡¥x»yªº¥¿½Tµo­µ¡C¡u¹q¸£»²§Uµo­µ°V½m»Pµû¤À¡v¬O»y­µ¿ëÃѪº¤@­Ó·s¿³¬ã¨s»PÀ³¥Î»â°ì¡A¬ÛÃöªº¤åÄm»P³ø§i¤éº¥Â×´I¡A³nÅéÀ³¥Î¤]¶V¨Ó¶V¦h¡A¦ý¤´¥¼¨£»P¥x»y¬ÛÃöªºÀ³¥Î¡C§Ú­Ì¦b¥ý«eªº²£·~¦X§@­pµe¤¤¡A¤w¸g³°Äò§¹¦¨¤F¡uµØ»y»y­µµû¤À¡v¡B¡u­^»y»y­µµû¤À¡v¡B¡u¤é»y»y­µµû¤À¡vµ¥¨t²Î¡A¦Ó¥B¤]¶i¦æ¬ÛÃöªº§Þ³NÂಾ¤Î°Ó«~¤Æ¡A¬ãµo¤Î²£·~¦X§@ªº¸gÅç¬Û·íÂ×´I¡C¦b¦¹­pµe¤¤¡A§Ú­Ì§Æ±æµ²¦X¨Ã¤Þ¥Î¨ä¥L¤l­pµe¥D«ù¤H¦b¥x»y»y¨¥³B²zªº¦h¦~¬ã¨s¦¨ªG¡A¹ê»Ú²£¥X¤@­Ó±µªñ¥«³õ»Ý¨Dªº¡u¹q¸£»²§U¥x»yµo­µµû¤À¨t²Î¡v¡A¦]¦¹¦¹­pµeªº¨Ï©R°£¤F¦b©ó¯à°÷¥H»y­µ¬ì§Þ¨Ó«P¶i¼Æ¦ì¾Ç²ß²£·~¤É¯Å¥~¡A§ó¨ã¦³¥»¤g¤å¤Æ±À¼sªº²`¤@¼h·N¸q¡C

  48. °ò©ó¼Ò¦¡ÃѧO¤èªk¶i¦æ¹q¾¹¯Ó¯à¯S¼x¤ÀªR

    • ­^¤å¦WºÙ: On the Use of Pattern Recognition Methods for Household Appliance Modeling Based on Readings of Electricity Meters
    • ­pµe½s¸¹:
    • ¥D«ù¤H: ±i´¼¬P
    • ¸É§U³æ¦ì: ¸êµ¦·|
    • ­pµe°õ¦æ´Á¶¡: 2010/3/1 to 2010/12/31
    • ÃöÁäµü: µL½u·P´úºô¸ôÀ³¥Î¨t²Î¡B¼Ò«¬ÃѧO¡B¾÷¾¹¾Ç²ß¡B¯S¼xÂ^¨ú¡B°T¸¹³B²z
    • ºK­n²¤¶:
      ¥»­pµe¥D­n±´°Q·P´úºô¸ôÀ³¥Î¨t²Î¤W¤@¶µ­«­nªºÄ³ÃD¡G¡u¦p¦ó³B²z²V¦X¦¡·P´ú¸ê®Æ¡v¥H¤Î¡u¦p¦ó¤ÀªR·P´ú¸ê®Æ¶Ç¹Fªº¸ê°T¡v¡C¥Ñ©ó·P´úºô¸ô©Ò§G«Øªº·P´ú¤¸¥ó¤§¶¡³sµ²¤è¦¡¤£¦P¡A·P´ú¤¸¥ó©Ò±½ºË¨ìªº¸ê®Æ¥i¯à¥]§t¦hºØ¤£¦Pµo°eºÝªº¸ê°T¡A³y¦¨·P´ú¤¸¥ó±µ¦¬¨ìªº¸ê°T¬Û·í½ÆÂø¡A¦]¦¹¥i¯à¾D¹JºÊ´ú¹ï¶HÃø¥H¤ÀªR¨ä¦æ¬°¤§°ÝÃD¡C¨Ò¦p¡A¥H¤@¯ë®a®xªº¥Î¹q¶q¶q´ú¬°¨Ò¡A¦]¬°¹qªí°O¿ý¤F¹q¤O¤½¥q¿é¤J¨ì¦¹®a®xªºÁ`¥Î¹q¥\²v¡AÁöµM·í¥Î¹q®É¡A¥i¬Ý¨ì»öªí¤£°±Âà°Ê¡A¥i¬O¦hºØ¹q¾¹¦P®É¹BÂà®É¡A«oµLªk±oª¾¦U­Ó¹q¾¹ªº¥Î¹q±¡§Î¡C¦pªG¯à°w¹ï³o¨Ç¸ê®Æ¶i¦æ¤ÀªR¡A±N¯à°÷ÁA¸Ñ¦U­Ó¹q¾¹ªº­Ó§O¹B§@ªpª¬¡A¨Ò¦p¬O§_¥¿¦b¤u§@©Î¬O³B¦b«Ý¾÷ª¬ºA¡A¦]¦¹¦p¦ó¤ÀªR²V¦X¤j¶q¤£¦P¨Ó·½ªº·P´ú¸ê®Æ¬O¤@¶µ­«­nªºÄ³ÃD¡C ¥»­pµe±N°w¹ï®a®x¥Î¹q±¡§Î¶i¦æ¤ÀªR¡AÁA¸Ñ¹q¤O­pªºªi§ÎÅܤƹïÀ³¹q¾¹¹B§@ªºÃö«Y¡C
      ¥»­pµe¤¤©Ò¬ãµo¤§§Þ³N±N¼¶¼g¦¨³ø§i»P´£¨Ñ¤ÀªR¤j¶q·P´úºô¸ô¸ê®Æªº´ú¸Õ¼Æ¾Ú¡A´£¨Ñ¸êµ¦·|¦b¶}µo¨ä¥L·P´úºô¸ôÀ³¥Î¨t²Î¸ê®Æ¤ÀªR§PŪ¤§°Ñ¦Ò¡C©Ò¬ãµo¤§§Þ³N±N¥Ó½Ð¤@¶µ±M§Q¡A¨Ã§ë½Z¤G½g¬ÛÃö½×¤å¡C

  49. ­ó°Û·j´M§Þ³N

    • ­^¤å¦WºÙ: Techniques for Query by Singing/Humming
    • ­pµe½s¸¹:
    • ¥D«ù¤H: ±i´¼¬P
    • ¸É§U³æ¦ì: ¸êµ¦·|
    • ­pµe°õ¦æ´Á¶¡: 2010/2/1 to 2010/12/31
    • ÃöÁäµü: »y­µ¿ëÃÑ¡B»y­µ·j´M¡B±Û«ß¿ëÃÑ¡B­ó°Û·j´M
    • ºK­n²¤¶:
      ¥»¦X§@¬ã¨s­pµeÀÀ¤Þ¤J­ó°Û·j´M§Þ³N¡A¥Î¥H·f°t©ú¬P©Î¤å³Ð³y«¬¤§¹q¤l¤½¥J¡C¨Ï¥ÎªÌ¥i³z¹L»y­µ¿é¤J­µ¼Ö©Î°t¼Öªº¤ù¬q¡A¥Ñ¨t²Î¹ï¨Ï¥ÎªÌªºÁn­µ¶i¦æ«eºÝ³B²z¡A¨ú±oÁn­µ¯S¼x«á¡A³z¹Lºq¦±¥D±Û«ß¸ê®Æ®w´M§ä¨Ã¿ëÃÑ¥X¹ïÀ³ªººq¦±¡A¦A±N¹ïÀ³¤§­µ¼Ö©Î°t¼ÖÀɮסA·f°t»y­µ«ü¥O§Þ³N¡A¦^°e¦Ü¹q¤l¤½¥J©M¹q¸£¿Ã¹õ¶Ç¹Fµ¹¨Ï¥ÎªÌ¡A©ÎªÌ°eµ¹¨Ï¥ÎªÌ«ü©wªºªB¤Í¡A¥H¼W¥[¹q¤l¤½¥Jªº®T¼Ö©Ê¤Î¤¬°Ê©Ê¡C

  50. ¾A¥Î©ó´O¤J¦¡¨t²Îªº¹q¸£»²§U¤f»¡µØ»yµo­µ½m²ß¨t²Î

    • ­^¤å¦WºÙ: Computer Aided Spoken Chinese Pronunciation Practice System for Embedded Systems
    • ­pµe½s¸¹:
    • ¥D«ù¤H: ±i´¼¬P
    • ¸É§U³æ¦ì: °ê¬ì·| & Üg«ä¬ì§Þ
    • ­pµe°õ¦æ´Á¶¡: 2009/11/1 to 2010/10/31
    • ÃöÁäµü:
    • ºK­n²¤¶:

  51. ¥Î©ó­µ°T­µ¼Öªº­pºâťı³õ´º¤ÀªR¡]3/3¡^

    • ­^¤å¦WºÙ: Computational Auditory Scene Analysis for Audio Music
    • ­pµe½s¸¹: NSC 96-2628-E-007 -141 -MY3
    • ¥D«ù¤H: ±i´¼¬P
    • ¸É§U³æ¦ì: °ê¬ì·|
    • ­pµe°õ¦æ´Á¶¡: 2009/8/1 to 2010/7/31
    • ÃöÁäµü: Computational Auditory Scene Analysis, Music Information Retrieval, Audio Music Analysis
    • ºK­n²¤¶:
      ©Ò¿×¡u­pºâťı³õ´º¤ÀªR¡v¡]Computational Auditory Scene Analysis¡A²ºÙCASA¡^ ªº¥Ø¼Ð¡A´N¬O­n±N³æÁn¹DªºÁn­µ¡]¥i¯à¥Ñ¦h­Ó­µ·½©Ò²£¥Í¡^¡A¸g¥Ñ¹q¸£ªº¦Û°Ê­pºâ¡A±Ä ¥Î¦UºØ¤èªk¡]¨Ò¦p§Ö³t³Å¥ß¸­Âà´«¡B°ÊºA³W¹º¡B¾÷¾¹¾Ç²ßµ¥¡^¡A¨ÃÀ³¥Î§Ú­Ì¹ï­µ·½©Ò¨ã ¦³ªº¦UºØª¾ÃÑ¡A¨Ó©â¨ú¥X³o¨Ç­µ·½ªºÁn­µ¡A¥H«K¶i¦æ¤U¤@¨Bªº³B²z¡C¦Û±qBregman ¦b 1990 ¦~´£¥XAuditory Scene Analysis ªº·§©À«á¡A¬ÛÃöªº¬ã¨s«ùÄò¤£Â_¡A¦ý³£ª`­«¦b ¤@¯ë»y­µªº³B²z¡Aª½¨ìªñ´X¦~¨Ó¡A­µ¼Ö¸g¥Ñºô»Úºô¸ô¤j¶q¶Ç¼½¡A¬ÛÃöªº¤ÀªR»PÀ˯Á¤] ¶V¨Ó¶V­«­n¡A¦]¦¹CASA ¦b­µ°T­µ¼Ö¡]Audio Music¡^¤è­±ªº¬ã¨s©MÀ³¥Î¤]¦b³o´X¦~¶} ©lµÞªÞ¡C¥»­pµe±N¾ã¦X¥»¹êÅç«Ç¶}µo¦h¦~ªº­µ°T³B²z§Þ³N¡]¥]§t±Û«ß¿ëÃÑ¡B»y­µ¿ëÃÑ¡B »y­µ»PºqÁn¦X¦¨¡B±j¶´¦¡­µ°ª°lÂÜ¡B»y­µÂà´«µ¥¡^¡A¸g¥ÑCASA ªº¬[ºc¨ÓÀ³¥Î©ó­µ°T­µ ¼Ö¡A§Æ±æ«Ø¥ß¦³®Äªº¤ÀªR¼Ò¦¡»P¤èªk¡A¯à°÷¹ï­µ°T­µ¼Ö¶i¦æ¤ÀªR»P³B²z¡C­pµeªº¥D­n ¥Ø¼Ð¡A¬O§Æ±æ°w¹ï¤@¯ë¬y¦æ­µ¼Ö¡A°µ¨ì¤U¦C´XÂI¡G
      1. ¹ï­µ°T­µ¼Ö¶i¦æ¥D±Û«ßªº­µ°ª°lÂÜ¡C
      2. §PÂ_ºqÁn¦s¦bªº¦ì¸m¡C
      3. ¥Ñ­µ°T­µ¼Ö©â¨ú³æ­µºqÁn¡C
      4. ¦h­«­µ°ª°lÂÜ¡C
      5. ©â¨ú¨ä¥L³æ­µ¼Ö¾¹ªºÁn­µ¡]¨Ò¦p¹ªÁn¡^¡C
      ¸g¥Ñ³o¨Ç¤ÀªR¡A§Ú­Ì¥i¥H¹ï­µ°T­µ¼Ö¶i¦æ§óºë±Kªº¤ÀÃþ»PÀ˯Á¡A¬ÛÃöªºÀ³¥Î«h¦³¡G
      1. ­µ°T­µ¼Öªº¦Û°Ê¤ÀÃþ»PÀ˯Á
      2. ­µ°T­µ¼Öªº­ó°ÛÀ˯Á
      3. ­µ°T­µ¼Öªº¸`©ç°lÂÜ
      4. ­µ°T­µ¼Öªººqµü¦P¨BÅã¥Ü
      5. ­µ°T­µ¼Öªº±¡ºü¤º®e¤ÀªR

  52. ¥xÆW¦Û¥D³B²z¾¹Android¥­¥x²`¯Ñ­pµe

    • ­^¤å¦WºÙ: Deep Cultivation for Taiwan's Independent Processor for Android Platforms
    • ­pµe½s¸¹:
    • ¥D«ù¤H: §õ¬F±X
    • ¸É§U³æ¦ì: ¸gÀÙ³¡¾Ç¬ã­pµe
    • ­pµe°õ¦æ´Á¶¡: 2009/6/1 to 2010/5/31
    • ÃöÁäµü:
    • ºK­n²¤¶:

  53. °Û§@­Ñ¨Î¦³Án®Ñ¹q¤l¤½¥J­pµe

    • ­^¤å¦WºÙ: Excellent Singing and Reading Electronic Figures for Audiobooks
    • ­pµe½s¸¹:
    • ¥D«ù¤H: ª÷¥ò¹F
    • ¸É§U³æ¦ì: ¸gÀÙ³¡¾Ç¬ã­pµe
    • ­pµe°õ¦æ´Á¶¡: 2009/6/1 to 2010/5/31
    • ÃöÁäµü:
    • ºK­n²¤¶:

  54. ¥H»yªÌ¿ëÃѬ°°ò¦¤§´¼¼z«¬¤H¾÷¤¶­±

    • ­^¤å¦WºÙ: Intelligent Man-machine Interface based on Speaker Recognition
    • ­pµe½s¸¹:
    • ¥D«ù¤H: ±i´¼¬P
    • ¸É§U³æ¦ì: ¸êµ¦·|
    • ­pµe°õ¦æ´Á¶¡: 2009/3/1 to 2009/12/31
    • ÃöÁäµü: »yªÌ¿ëÃÑ¡B»y­µ°T¸¹³B²z¡B¥Íª«»{ÃÒ¡B´¼¼z«¬¤H¾÷¤¶­±
    • ºK­n²¤¶:
      ¨Ï¥Î¤HªºÁn­µ¨Ó¶i¦æ¥Íª«»{ÃÒ¡A¬O¤@¶µ«D±`¦³§l¤Þ¤Oªº§Þ³Nµo®i¤è¦V¡A¦]¬°Án­µªºÂ^¨ú«D±`®e©ö¡A¤£»Ý­n¯S®íªºµwÅé¡A¦P®É¤]¤£·|³y¦¨¨Ï¥ÎªÌªº¾á¤ß®`©È¡A«I¤J©Ê¸û§C¡C¦ý¬OÁn­µ¤]®e©ö¨ü¨ì¥~¬ÉÂø°Tªº¤zÂZ¡A¦P®É¤]®e©ö¨ü¨ì»¡¸ÜªÌ¥»¨­ªº¨­Å鱡ªp©Ò¼vÅT¡A³o¬O¦¹§Þ³Nªº¯ÊÂI¡C¥Ñ©ó¹q¸£¹Bºâ³t«×ªº¬ð­¸²r¶i¡A¦]¦¹»yªÌ¿ëÃѪºµo®i¤]º¥º¥¬ð¯}³o¨ÇªùÂe¡A³vº¥Åܦ¨¹ê»Ú¥i¥Îªº§Þ³N¡C¥»­pµe±N¶}µo¤@®M¤å¥»¬ÛÃöªº»yªÌ¿ëÃѨt²Î¡A¥H«K¹êÃÒ¦¹§Þ³Nªº¥i¥Î«×»P¦¨¼ô«×¡A¨Ã³]ªk§JªA»yªÌ¿ëÃѦb¹ê¥Î¤W·|¸I¨ìªº°ÝÃD¡A¥H«KÀu¤Æ¾ãÅé¨t²Î¡A¹F¨ì¥i¹ê»Ú¶i¦æ°Ó·~¥Î³~ªº¥Ø¼Ð¡C

  55. ±q»y­µ¹ï¸Ü¶i¦æ±¡ºü¿ëÃÑ

    • ­^¤å¦WºÙ: Emotion Detection from Spoken Dialog
    • ­pµe½s¸¹:
    • ¥D«ù¤H: ±i´¼¬P
    • ¸É§U³æ¦ì: ¸êµ¦·|
    • ­pµe°õ¦æ´Á¶¡: 2009/3/1 to 2009/12/31
    • ÃöÁäµü: »y­µ°T¸¹³B²z¡B¼Ë¦¡¿ë»{¡BÃöÁäµüÂ^¨ú
    • ºK­n²¤¶:
      ¤@­Ó¤Hªº±¡ºü¥~¦bªí²{¡A¥D­n¸g¥ÑªÏÅé°Ê§@¡BÁy³¡ªí±¡¡B¤f»¡»y¨¥µ¥¤TºØ¤è¦¡¨Óªí²{¡A¦Ó¨ä¤¤²o¯A¨ìªº¸ê®Æ¶q¡]«ü¥i¥Hª½±µ°e¤J¹q¸£¤ÀªRªº¸ê®Æ¡^¡A¤S¥H¤f»¡»y¨¥¬°³Ì¤Ö¡]¥u¦³¤@ºûªº­µ°T¸ê®Æ¡^¡A¦ý¤f»¡»y¨¥«o¥i¥Hªí¹F¥X«Ü²Ó¿°ªº±¡ºü¡A¦]¦¹¦b±¡ºü¿ëÃѪº¬ã¨s¤è­±¡A»y­µ¹ï¸Ü´NÅܦ¨¤@­Ó«D±`­«­nªº¬ã¨s½u¯Á¡A³o¤]¬Oªñ´X¦~¨Ó«D±`¼öªùªº¬ã¨sÃD§÷¡C

      ¦bÀ³¥Î¤è­±¡A¥H¡u»y­µ¶i¦æ±¡ºü¿ëÃÑ¡v¤]¦³¤£¦Pªº­±¹³©MµÛ¤OÂI¡A¨Ò¦p¡A¦b¤¬°Ê¹q¤lÃdª«ªºÀ³¥Î¤W¡A§Ú­Ì¥i¥H¸g¥Ñ¥D¤Hªº»y­µ¨Ó°»´ú¨ä±¡ºü¡A¨Ã¶i¦Ó±À½×³Ì¨Îªº¦^¸Ü»P¤¬°Ê¤è¦¡¡A¥H«K´£¨ÑÅé¶K¤J·LªºªA°È¡AÅý¥D¤H¦³¶K¤ßªº·Pı¡C¦b°Ó·~À³¥Î¤è­±¡A§Ú­Ì¥i¥H¸g¥Ñ«È¤á¦b«ÈªA±M½uªº»y­µ¹ï¸Ü¨Ó§Pª¾¨ä±¡ºü¡A¨Ã¶i¦ÓÁA¸Ñ«ÈªA¤H­û¦b¦w¼¾«È¤áªº¥\¤O¤Îªí²{¡C¦¹¥~¡A¦b¤@¯ë¤â¾÷³q¸ÜªºÀ³¥Î¡A§Ú­Ì¤]¥i¥H¶}³q¡u¤ß¤ß¬Û¬M«ü¼Æ¡vªA°È¡A¥H»y­µ¨Ó§PÂ_¨â¤H¹ï¸Üªº´r®®«×¡C

      °£¤F»y­µ¤§¥~¡A¤¬°Ê¹q¤lÃdª«¤]¯à°÷¸g¥ÑÄá¼v¾÷»´©ö¦a¨ú±o¨ì¥D¤HªºÁy³¡ªí±¡©M°Ê§@µ¥¡A¦]¦¹¤~¯à°÷§ó¶i¤@¨B¦a¡u¹î¨¥Æ[¦â¡v¡C¸g¥Ñ³oºØ¦h¼Ò¦¡ªº±¡·P­pºâ¡]Multi-modal Affective Computing¡^¡A¤~¯à°÷§ó·Ç½T¦a§PÂ_¤@­Ó¤Hªº±¡ºüª¬ºA¡A³o¤]¬O¥»­pµeªº¾ã¦X¬ã¨s­«ÂI¡C

  56. ´O¤J¦¡¦h®Ö¤ß½sĶ¾¹»PÀ³¥Î³nÅ饭¥x¬ãµo­pµe

    • ­^¤å¦WºÙ: R & D for Embedded Multi-core Compiler and Application Software Platform
    • ­pµe½s¸¹:
    • ¥D«ù¤H: §õ¬F±X
    • ¸É§U³æ¦ì: ²MµØ¤j¾Ç
    • ­pµe°õ¦æ´Á¶¡: 2009/3/1 to 2010/12/31
    • ÃöÁäµü: ´O¤J¦¡¨t²Î
    • ºK­n²¤¶:

  57. ´O¤J¦¡²§¦h®Ö¤ß¨t²Î§Þ³N¬ãµo3¦~­pµe(²Ä2´Á)

    • ­^¤å¦WºÙ: Embedded heterogeneous multi-core system technology research and development 3-year plan
    • ­pµe½s¸¹:
    • ¥D«ù¤H: ±i´¼¬P
    • ¸É§U³æ¦ì: ¸gÀÙ³¡¬ì±M­pµe
    • ­pµe°õ¦æ´Á¶¡: 2008/11/1 to 2010/10/31
    • ÃöÁäµü:
    • ºK­n²¤¶:

  58. IntelÁp¦X¬ãµo­pµe

    • ­^¤å¦WºÙ: Intel Joint Research and Development Program
    • ­pµe½s¸¹:
    • ¥D«ù¤H: ±i´¼¬P
    • ¸É§U³æ¦ì: Intel
    • ­pµe°õ¦æ´Á¶¡: 2008/8/1 to 2009/7/31
    • ÃöÁäµü:
    • ºK­n²¤¶:
      In this subproject, we shall devote to the deployment of speech technology for innovative and user-aware and location-aware MID applications, including the following potential items:
      1. Voice commands for MID applications
        Using simple voice commands for invoking MID applications is likely to be a feasible way of enable speech technology for natural user interface.
      2. Speaker identification/verification
        We can use voiceprint to identify a user and then adopt personalized MID settings. Such user-aware scenario will definitely improve user experience. Moreover, we can also apply speaker verification for authentication on MID.
      3. Computer-assisted pronunciation training (CAPT)
        Spoken language learning is a newly developed application area in speech technology. We can implement CAPT on MID for spoken language learning, including English and Mandarin.
      4. Speech-based retrieval of location-aware information
        Speech-enable interface for retrieving location-aware information is a practical application for everyday¡¦s needs. In particular, we can focus on specific domains, such as travel and sightseeing. Most commonly used keywords for such domain are ¡§restaurant¡¨, ¡§rest room¡¨, ¡§department store¡¨, ¡§bus stop¡¨, and so on. Once our system receives the keyword, it should retrieve the corresponding information based on the geographic location to give location-aware results that can best suit the user¡¦s needs.

  59. ¥Î©ó­µ°T­µ¼Öªº­pºâťı³õ´º¤ÀªR¡]2/3¡^

    • ­^¤å¦WºÙ: Computational Auditory Scene Analysis for Audio Music
    • ­pµe½s¸¹: NSC 96-2628-E-007 -141 -MY3
    • ¥D«ù¤H: ±i´¼¬P
    • ¸É§U³æ¦ì: °ê¬ì·|
    • ­pµe°õ¦æ´Á¶¡: 2008/8/1 to 2009/7/31
    • ÃöÁäµü: Computational Auditory Scene Analysis, Music Information Retrieval, Audio Music Analysis
    • ºK­n²¤¶:
      ©Ò¿×¡u­pºâťı³õ´º¤ÀªR¡v¡]Computational Auditory Scene Analysis¡A²ºÙCASA¡^ ªº¥Ø¼Ð¡A´N¬O­n±N³æÁn¹DªºÁn­µ¡]¥i¯à¥Ñ¦h­Ó­µ·½©Ò²£¥Í¡^¡A¸g¥Ñ¹q¸£ªº¦Û°Ê­pºâ¡A±Ä ¥Î¦UºØ¤èªk¡]¨Ò¦p§Ö³t³Å¥ß¸­Âà´«¡B°ÊºA³W¹º¡B¾÷¾¹¾Ç²ßµ¥¡^¡A¨ÃÀ³¥Î§Ú­Ì¹ï­µ·½©Ò¨ã ¦³ªº¦UºØª¾ÃÑ¡A¨Ó©â¨ú¥X³o¨Ç­µ·½ªºÁn­µ¡A¥H«K¶i¦æ¤U¤@¨Bªº³B²z¡C¦Û±qBregman ¦b 1990 ¦~´£¥XAuditory Scene Analysis ªº·§©À«á¡A¬ÛÃöªº¬ã¨s«ùÄò¤£Â_¡A¦ý³£ª`­«¦b ¤@¯ë»y­µªº³B²z¡Aª½¨ìªñ´X¦~¨Ó¡A­µ¼Ö¸g¥Ñºô»Úºô¸ô¤j¶q¶Ç¼½¡A¬ÛÃöªº¤ÀªR»PÀ˯Á¤] ¶V¨Ó¶V­«­n¡A¦]¦¹CASA ¦b­µ°T­µ¼Ö¡]Audio Music¡^¤è­±ªº¬ã¨s©MÀ³¥Î¤]¦b³o´X¦~¶} ©lµÞªÞ¡C¥»­pµe±N¾ã¦X¥»¹êÅç«Ç¶}µo¦h¦~ªº­µ°T³B²z§Þ³N¡]¥]§t±Û«ß¿ëÃÑ¡B»y­µ¿ëÃÑ¡B »y­µ»PºqÁn¦X¦¨¡B±j¶´¦¡­µ°ª°lÂÜ¡B»y­µÂà´«µ¥¡^¡A¸g¥ÑCASA ªº¬[ºc¨ÓÀ³¥Î©ó­µ°T­µ ¼Ö¡A§Æ±æ«Ø¥ß¦³®Äªº¤ÀªR¼Ò¦¡»P¤èªk¡A¯à°÷¹ï­µ°T­µ¼Ö¶i¦æ¤ÀªR»P³B²z¡C­pµeªº¥D­n ¥Ø¼Ð¡A¬O§Æ±æ°w¹ï¤@¯ë¬y¦æ­µ¼Ö¡A°µ¨ì¤U¦C´XÂI¡G
      1. ¹ï­µ°T­µ¼Ö¶i¦æ¥D±Û«ßªº­µ°ª°lÂÜ¡C
      2. §PÂ_ºqÁn¦s¦bªº¦ì¸m¡C
      3. ¥Ñ­µ°T­µ¼Ö©â¨ú³æ­µºqÁn¡C
      4. ¦h­«­µ°ª°lÂÜ¡C
      5. ©â¨ú¨ä¥L³æ­µ¼Ö¾¹ªºÁn­µ¡]¨Ò¦p¹ªÁn¡^¡C
      ¸g¥Ñ³o¨Ç¤ÀªR¡A§Ú­Ì¥i¥H¹ï­µ°T­µ¼Ö¶i¦æ§óºë±Kªº¤ÀÃþ»PÀ˯Á¡A¬ÛÃöªºÀ³¥Î«h¦³¡G
      1. ­µ°T­µ¼Öªº¦Û°Ê¤ÀÃþ»PÀ˯Á
      2. ­µ°T­µ¼Öªº­ó°ÛÀ˯Á
      3. ­µ°T­µ¼Öªº¸`©ç°lÂÜ
      4. ­µ°T­µ¼Öªººqµü¦P¨BÅã¥Ü
      5. ­µ°T­µ¼Öªº±¡ºü¤º®e¤ÀªR

  60. Ápµo¬ì´O¤J¦¡¨t²Î§Þ³N¬ã¨s¤Î¤H¤~°ö¨|­pµe¡]²Ä¥|¤l­pµe¡Gµø°T¤Î»y­µÀ³¥Î¶}µo¡^

    • ­^¤å¦WºÙ: MediaTek Embedded System Technology Research and Talent Cultivation Program (Fourth Sub-Program: Video and Voice Application Development)
    • ­pµe½s¸¹:
    • ¥D«ù¤H: ±i´¼¬P
    • ¸É§U³æ¦ì: Ápµo¬ì
    • ­pµe°õ¦æ´Á¶¡: 2008/8/1 to 2009/7/31
    • ÃöÁäµü:
    • ºK­n²¤¶:

  61. Tri-toneªº³sÄòÁn½Õ¶ì¼Ò¤Î°»¿ù§Þ³N

    • ­^¤å¦WºÙ: Tri-tone Based Continuous Tone Modeling and Analysis
    • ­pµe½s¸¹:
    • ¥D«ù¤H: ±i´¼¬P
    • ¸É§U³æ¦ì: ¸êµ¦·|
    • ­pµe°õ¦æ´Á¶¡: 2008/3/1 to 2008/11/30
    • ÃöÁäµü: ¹q¸£»²§Uµo­µ½m²ß¡BÁn½Õ¿ëÃÑ¡BÁôÂæ¡°¨¥i¤Ò¼Ò«¬¡B­µ°ª°lÂÜ
    • ºK­n²¤¶:
      ¥Ñ©ó¹q¸£³t«×ªº¼W¶i¥H¤Î»y­µ¬ì§Þªººt¶i¡A»y­µ¿ëÃѪºÀ³¥Î»â°ì¤w¸g±q³æ¯Âªº¤H¾÷¤¶­±Âà¨ì½ÆÂø«×§ó°ª¡B¥Î³~§ó¼sªxªº¹q¸£»²§Uµo­µ°V½m»Pµû¤À¡C¥HµØ»y¦Ó¨¥¡Aµû¤Àªº¼Ð·Ç°£¤F­µ¦â¤§¥~¡AÁÙ¥]§tÁn½Õ¡A¦]¬°µØ»y¬O©Ò¿×ªºtonal language¡A¤×¨ä¬O¹ï©ó¥~°ê¤H¦Ó¨¥¡A¥¿½TªºÁn½Õ»·¤ñ­µ¦â¨Ó±o§xÃø¡A¦]¦¹¥»­pµe±N±´°Q¦p¦ó¨Ï¥Î»y­µ¬ÛÃö§Þ³N¡A¨Ó¶i¦æÁn½Õªº¶ì¼Ò»P¤ÀªR¡A¨Ã¹Á¸Õ¨Ï¥Î¥»­pµe©Ò¶}µoªºÁn½Õ¿ëÃÑ©ó¾ã¦X©ÊªºµØ»y¹q¸£»²§U¾Ç²ß¨t²Î¤§¤¤¡A¥H´£°ª¹q¸£»²§Uµo­µ°V½m¦bµØ»y¤è­±ªº¥þ­±©Ê»P¥i¥Î©Ê¡C

  62. »y­µ¿ëÃѨt²Î¶}µo

    • ­^¤å¦WºÙ: Speech Recognition System Development
    • ­pµe½s¸¹:
    • ¥D«ù¤H: ±i´¼¬P
    • ¸É§U³æ¦ì: ¤¤¬ì°|
    • ­pµe°õ¦æ´Á¶¡: 2008/2/1 to 2008/11/30
    • ÃöÁäµü:
    • ºK­n²¤¶:

  63. ¥Î©ó­µ°T­µ¼Öªº­pºâťı³õ´º¤ÀªR¡]1/3¡^

    • ­^¤å¦WºÙ: Computational Auditory Scene Analysis for Audio Music
    • ­pµe½s¸¹: NSC 96-2628-E-007 -141 -MY3
    • ¥D«ù¤H: ±i´¼¬P
    • ¸É§U³æ¦ì: °ê¬ì·|
    • ­pµe°õ¦æ´Á¶¡: 2007/8/1 to 2008/7/31
    • ÃöÁäµü: Computational Auditory Scene Analysis, Music Information Retrieval, Audio Music Analysis
    • ºK­n²¤¶:
      ©Ò¿×¡u­pºâťı³õ´º¤ÀªR¡v¡]Computational Auditory Scene Analysis¡A²ºÙCASA¡^ ªº¥Ø¼Ð¡A´N¬O­n±N³æÁn¹DªºÁn­µ¡]¥i¯à¥Ñ¦h­Ó­µ·½©Ò²£¥Í¡^¡A¸g¥Ñ¹q¸£ªº¦Û°Ê­pºâ¡A±Ä ¥Î¦UºØ¤èªk¡]¨Ò¦p§Ö³t³Å¥ß¸­Âà´«¡B°ÊºA³W¹º¡B¾÷¾¹¾Ç²ßµ¥¡^¡A¨ÃÀ³¥Î§Ú­Ì¹ï­µ·½©Ò¨ã ¦³ªº¦UºØª¾ÃÑ¡A¨Ó©â¨ú¥X³o¨Ç­µ·½ªºÁn­µ¡A¥H«K¶i¦æ¤U¤@¨Bªº³B²z¡C¦Û±qBregman ¦b 1990 ¦~´£¥XAuditory Scene Analysis ªº·§©À«á¡A¬ÛÃöªº¬ã¨s«ùÄò¤£Â_¡A¦ý³£ª`­«¦b ¤@¯ë»y­µªº³B²z¡Aª½¨ìªñ´X¦~¨Ó¡A­µ¼Ö¸g¥Ñºô»Úºô¸ô¤j¶q¶Ç¼½¡A¬ÛÃöªº¤ÀªR»PÀ˯Á¤] ¶V¨Ó¶V­«­n¡A¦]¦¹CASA ¦b­µ°T­µ¼Ö¡]Audio Music¡^¤è­±ªº¬ã¨s©MÀ³¥Î¤]¦b³o´X¦~¶} ©lµÞªÞ¡C¥»­pµe±N¾ã¦X¥»¹êÅç«Ç¶}µo¦h¦~ªº­µ°T³B²z§Þ³N¡]¥]§t±Û«ß¿ëÃÑ¡B»y­µ¿ëÃÑ¡B »y­µ»PºqÁn¦X¦¨¡B±j¶´¦¡­µ°ª°lÂÜ¡B»y­µÂà´«µ¥¡^¡A¸g¥ÑCASA ªº¬[ºc¨ÓÀ³¥Î©ó­µ°T­µ ¼Ö¡A§Æ±æ«Ø¥ß¦³®Äªº¤ÀªR¼Ò¦¡»P¤èªk¡A¯à°÷¹ï­µ°T­µ¼Ö¶i¦æ¤ÀªR»P³B²z¡C­pµeªº¥D­n ¥Ø¼Ð¡A¬O§Æ±æ°w¹ï¤@¯ë¬y¦æ­µ¼Ö¡A°µ¨ì¤U¦C´XÂI¡G
      1. ¹ï­µ°T­µ¼Ö¶i¦æ¥D±Û«ßªº­µ°ª°lÂÜ¡C
      2. §PÂ_ºqÁn¦s¦bªº¦ì¸m¡C
      3. ¥Ñ­µ°T­µ¼Ö©â¨ú³æ­µºqÁn¡C
      4. ¦h­«­µ°ª°lÂÜ¡C
      5. ©â¨ú¨ä¥L³æ­µ¼Ö¾¹ªºÁn­µ¡]¨Ò¦p¹ªÁn¡^¡C
      ¸g¥Ñ³o¨Ç¤ÀªR¡A§Ú­Ì¥i¥H¹ï­µ°T­µ¼Ö¶i¦æ§óºë±Kªº¤ÀÃþ»PÀ˯Á¡A¬ÛÃöªºÀ³¥Î«h¦³¡G
      1. ­µ°T­µ¼Öªº¦Û°Ê¤ÀÃþ»PÀ˯Á
      2. ­µ°T­µ¼Öªº­ó°ÛÀ˯Á
      3. ­µ°T­µ¼Öªº¸`©ç°lÂÜ
      4. ­µ°T­µ¼Öªººqµü¦P¨BÅã¥Ü
      5. ­µ°T­µ¼Öªº±¡ºü¤º®e¤ÀªR

  64. µØ»y¤å¤¬°Ê»y­µ±Ð¾Ç§Þ³N¬ãµo

    • ­^¤å¦WºÙ: Speech-based Dialog Technologies for Learning Mandarin Chinese
    • ­pµe½s¸¹:
    • ¥D«ù¤H: ±i´¼¬P
    • ¸É§U³æ¦ì: ¸êµ¦·|
    • ­pµe°õ¦æ´Á¶¡: 2007/5/1 to 2007/12/31
    • ÃöÁäµü: ¤¤¤å»y­µ¿ëÃÑ¡B¹q¸£»²§Uµo­µ°V½m¡B¹q¸£»²§U¤f»¡»y­µ¾Ç²ß
    • ºK­n²¤¶:
      ¥»­pµe±N±Ä¥Î¤¤¤å»y­µ¿ëÃÑ»P»y­µµû¤Àªº§Þ³N¡A¶}µo¤@®M¹q¸£»²§U¤f»¡¤¤¤åªº¾Ç²ß¨t²Î¡A¦P®É¦b PC ¥H¤Î´O¤J¦¡¨t²Î¡]PDA¡BPMP¡BSmartPhoneµ¥¤â«ù¦¡¸Ë¸m¡^¶i¦æ¨t²Î¹ê²{¡C¨Ï¥ÎªÌ¥u­n¹ï³Á§J­·Á¿¤@¥y¤¤¤å¡A¦¹¨t²Î§Y¥i®Ú¾Ú¿é¤J»y­µ¸ê®Æ¡A§ì¨ú­µ¦â¡]»y­µªº¤º®e¡^¡BÁn½Õ¡]°ê»y¥|­ÓÁn½Õªº¤ÀÃþ¡^¡BÃý«ß¡]¨C¤@­Ó­µªº®É¶¡ªøµu¡^¡B­µ¶q¡]¨C¤@­Ó­µªº±j«×¡^µ¥»y­µ¯S¼x¡AµM«á¶i¦æ©M¥Ø¼Ð»y¥y©Î¬O»y­µ¼Ò«¬ªº¤ñ¹ï¡A®Ú¾Ú¨ä¬Û¦ü«×¨Ó²£¥Í¤@­Ó¤¶©ó0©M100¤§¶¡ªº¤À¼Æ¡C©M«e¤@¦~ªº­pµe¤ñ°_¨Ó¡A¥»¦~©Ò¶}µoªº¨t²Î§óµÛ­«©ó²V²c­µªº°»´ú¡A¦]¦¹¥²¶·¹ï¾Ç²ßªÌªº¥À»y¶i¦æ¨Æ«e¤ÀªR¡A¥H¨t²Î¤Æªº¤è¦¡¨Ó§ä¥X¨Ï¥ÎªÌ¦b¾Ç²ßº~»y®É¡A¥i¯à²£¥Íªº²V²c­µ¡]³q±`¬O¥Ñ¥À»y¤zÂZ©Òµo¥Í¡^¡^¡A¨Ã¦b¹ê»ÚÀ³¥Î®É¡A¯à°÷·Ç½T¦a°»´ú²V²c­µªºµo¥Í¨Ãµ¹¤©¨Ï¥ÎªÌµo­µ§ï¶iªº«Øij¡C¦¹¥~¡A§Ú­Ì¤]­n¥[±j¹ï¸Ü±¡¹Òªº½m²ß¡A¦]¦¹¥²¶·¹ï¹ï¸Ü±Ð§÷ªº·Ç³Æ¡A¦h¤U¥\¤Ò¡C

  65. ¥H»y­µ¿ëÃѤΦX¦¨¬°°ò¦ªº¤f»¡µØ»y¤å¹q¸£»²§U¾Ç²ß¨t²Î

    • ­^¤å¦WºÙ: A Spoken Mandarin Learning System Based on Speech Recognition and Synthesis
    • ­pµe½s¸¹: 95-3113-S-007-001-
    • ¥D«ù¤H: ±i´¼¬P
    • ¸É§U³æ¦ì: °ê¬ì·|
    • ­pµe°õ¦æ´Á¶¡: 2006/12/1 to 2007/11/30
    • ÃöÁäµü: µØ»y¤å±Ð¾Ç¡B¹q¸£»²§U»y¨¥¾Ç²ß¡B¹q¸£»²§Uµo­µ°V½m¡B»y­µ¿ëÃÑ¡B»y­µ¦X¦¨
    • ºK­n²¤¶:
      ¤@ºØ»y¨¥ªº¾Ç²ß¡A¥i¥H¤j²¤¤À¦¨¡uÅ¥¡B»¡¡BŪ¡B¼g¡v¥|¤j­±¦V¡A¨ä¤¤¡uÅ¥¡B»¡¡v¬O²o¯A¨ì¡u»y­µ¡vªº±µ¦¬©M²£¥Í¡A¦Ó¡uŪ¡B¼g¡v«h¬O²o¯A¨ì¡u¤å¦r¡vªº±µ¦¬»P²£¥Í¡C¥»­pµe±N¾ã¦X¥»¹êÅç«Ç¶}µo¦h¦~¥B¹ê»Ú¥i¦æªº»y­µ§Þ³N¡]¥]§t»y­µ¿ëÃÑ¡B»y­µ¦X¦¨¡B»y­µÂà´«µ¥¡^¡A¨Ó»²§U¾Ç²ßªÌ¦bµØ»y¤å¡uÅ¥¡B»¡¡v¤è­±ªº¾Ç²ß¡A¨Ã»P¤¸´¼¤j¾Çù»ñ¯]¦Ñ®v©Ò´£¥Xªº¡u§¹¾ã¦h¤¸µØ»y¤åºô¸ô¾Ç°|¡v­pµe¦X§@¡A¨Ï¥Î¸Ó­pµeªº°ê»y¤é³øªÀ¡B¹´±ÐÂù¶g¥Z¡B¬ü¬wµØ»y½Ò¥»¡B¤¤°ê¥j¨å¤å¾Çµ¥¬ÛÃö¯À§÷¡A¨Ã»P°ê»y¤é³øªÀ¦X§@¡A¨Ï¥Î¬ÛÃöªº±Ð§÷¡A¥Ø¼Ð¬O²£¥X¤@­Ó¹ê»Ú¥i¥Îªº¡u¹q¸£»²§U¤f»¡µØ»y¤åµo­µ½m²ß¤Îµû¤À¨t²Î¡v¡A¨ã¦³¤U¦C¤T¶µ¥\¯à¡G
      1. ¹ï¨Ï¥ÎªÌªº³æ¥yµØ»yµo­µ¶i¦æµû¤À¡Aµû¤À¼h¯Å¥]§t¨C¤@­Óª`­µ¡B¦r¡B¥y¡Aµû¤À¼Ð·Ç¥]§t­µ¦â¡BÃý«ß¡]¬y§Q«×¡^¡B­µ½Õ¡B­µ¶qµ¥¡A¨Ãµ¹¤©µo­µ§ï¶iªº«Øij¡C
      2. ¨Ï¥Î»y­µ¿ëÃѪº§Þ³N©M¨Ï¥ÎªÌ¶i¦æ¦UºØ±¡¹Ò¹ï¸Ü¡C
      3. ¥H»y­µ¦X¦¨ªº¤è¦¡¡A¨ÓÅý¨Ï¥ÎªÌ¶i¦æÅ¥¤O´úÅç¡C
      ¦¹¤T¶µ¥\¯à¡A¯à°÷Åý¨Ï¥ÎªÌ¸g¥Ñ»P¹q¸£ªº¤ÏÂФ¬°Ê¡A¶i¦æµo­µ½m²ß¡B±¡¹Ò¹ï¸Ü¤ÎÅ¥¤O´úÅç¡A¨Ã¸g¥Ñ¹q¸£ªº¦^õX¡]¥]§t¹ªÀy¡B»¡©ú¡BÁB¥¿¡^¡A¥[±j¦Û¤v¤f»¡µØ»y¤åªº¥¿½Tµo­µ¡A¹F¨ì¾Ç²ßµØ»y¤åªº¡uÅ¥¡B»¡¡vªº¥Ø¼Ð¡C

  66. Å¥¼g¨t²Î¤§»yªÌ½Õ¾A§Þ³N

    • ­^¤å¦WºÙ: Speaker Adaptation for an Embedded Dictation System
    • ­pµe½s¸¹:
    • ¥D«ù¤H: ±i´¼¬P
    • ¸É§U³æ¦ì: ÁÉ·L¬ì§Þ
    • ­pµe°õ¦æ´Á¶¡: 2006/11/1 to 2007/10/31
    • ÃöÁäµü: »yªÌ½Õ¾A¡B»y­µ¿ëÃÑ¡BÅ¥¼g¨t²Î
    • ºK­n²¤¶:
      ¥»­pµe¹Á¸Õ¨Ï¥Î»yªÌ½Õ¾A¤§¦U¶µ§Þ³N¡A§ï¶i¥Î©ó SmartPhone ¤§Å¥¼g¨t²Î¡A¥H«K¥Î©ó¤â¾÷¤§»y­µÂರTªº¥\¯à¡C

  67. ­µ¼Ö·j´Mªº¥[³t»P¿ëÃѲv´£¤É¡A¤Î¨ä¦b´O¤J¦¡¨t²Îªº¹ê§@»PÀ³¥Î¡]3/3¡^

    • ­^¤å¦WºÙ: Speedup Mechansim and Performance Enhancement for Music Information Retrieval, with Applications to Embedded Systems
    • ­pµe½s¸¹: 95-2221-E-007-220-
    • ¥D«ù¤H: ±i´¼¬P
    • ¸É§U³æ¦ì: °ê¬ì·|
    • ­pµe°õ¦æ´Á¶¡: 2006/8/1 to 2007/7/31
    • ÃöÁäµü: ­µ¼Ö¸ê°TÀ˯Á¡B¦h¼Ò¦¡¨Ï¥ÎªÌ¤¶­±¡B±Û«ß¿ëÃÑ¡B­µ°ª°lÂÜ¡B¦h´CÅé¸ê®Æ®wÀ˯Á¡B°ÊºA®É¶¡®Õ¥¿
    • ºK­n²¤¶:
      ¥Ñ©ó¥»­pµe¥D«ù¤H¹L¥h¼Æ¦~¦b°ê¬ì·|¡B±Ð¨|³¡¨ô¶V­pµe¤Î²M½«¬ì§Þ¤§¸g¶O¤ä´©¤U¡A¤w¸g¦¨¥\¦a«Ø¥ß¤F¤@­Ó¤j«¬ªº­µ¼ÖÀ˯Á¨t²Î¡A¦¬¶°¤F¶W¹L¨â¸U­ººq¦±¡AÅý¨Ï¥ÎªÌ¯à¥H­ó°Û¡B»y­µ»PºVÀ»µ¥¤£¦P¤è¦¡¨Ó¿ï¨úºô»Úºô¸ô¤Wªººq¦±¡C¬°¤F¯à°÷¥H§Ö³tªº¤èªk¹ï¤j¶q¸ê®Æ¶i¦æ¦³®Äªº¤ñ¹ï¡A§Ú­Ì°£¤F¦b¤ñ¹ï¤èªk¶i¦æ²`¤J¬ã¨s¥~¡A¤]¬[³]¤Fºô®æ¹Bºâ¡]Grid Computing¡^ªº¦øªA¾¹¥­¥x¡A´Á±æ¥H¥­¦æ¤Î¤À´²³B²zªº¤è¦¡¡A¼W¶i¨t²Îªº¥i¥Î©Ê¡C¦b¦¹ºØ¥]§t¤j¶q¸ê®Æ³B²z»P¤ñ¹ïªº¨t²Î¤¤¡A­l¥Í¤F³\¦h¬ÛÃö¬ã¨sªº¬ã¨sijÃD¡A¥»¬ã¨s­pµeÀÀ¥H¤T¦~®É¶¡¡A°w¹ï³o¨ÇijÃD¶i¦æ²`¤J±´°Q¡A³o¨ÇijÃD¤Î¬ÛÃö¬ã¨s®Éµ{¦C¥X¦p¤U¡G
      1. ²Ä¤@¦~¡G ¥H´£°ª¿ëÃѲv¬°¥D
        1. ­µ°ª°lÂÜ¡]Pitch Tracking¡^ªº¦UºØ·s¤èªkªº¹Á¸Õ¡A¥]§t Combination of Classifiers¡BACF/AMDF Map µ¥¡C
        2. °ÊºA®É¶¡®Õ¥¿¡]Dynamic Time Warping¡^ªº§ï¶i»P´ú¸Õ¡A¥]§t Local Path Constraints¡BPath Penalty¡BFlexible Starting Points µ¥¡C
      2. ²Ä¤G¦~¡G¥H¥[§Ö¿ëÃѳt«×¬°¥D
        1. ¡u¨â¶¥¬q¦¡¤ñ¹ïªk¡vªº±´°Q»P§ï¨}¡A¥]§t¥H¤£¦Pªº¤è¦¡¶i¦æ¨â¶¥¬q¤ñ¹ï¡A¨Ò¦p Linear Scaling ¥[ DTW (Dynamic Time Warping)¡B
        2. ¨ä¥L¦UºØ­µ¼ÖÀ˯Á¤èªkªº±´°Q»P¥[³t¡A¥]§t¦p¦óºë½T¦a­pºâ DTW ªº Lower Bound¡B¦p¦ó¥H FastMap ªº¤èªk¨Ó±Nºq¦±Âà´«¨ì°ª¬°ªÅ¶¡¤¤ªºÂIµ¥¡C
        3. ÁôÂæ¡°¨¥i¤Ò¼Ò«¬¡]Hidden Markov Models¡^ªº¹Á¸Õ»P§ï¶i
      3. ²Ä¤T¦~¡G¥H´O¤J¦¡¨t²Îªº²¾´Ó¬°¥D
        1. ­µ°ª°lÂÜ»PÀ˯Á¤èªkªººë¶i»P²¤Æ
        2. ´O¤J¦¡¨t²Îªº¹ê§@»PÀ³¥Î

  68. PDA¤¤¤å»y­µ¦X¦¨¨t²Î

    • ­^¤å¦WºÙ: Chinese TTS for PDA
    • ­pµe½s¸¹:
    • ¥D«ù¤H: ±i´¼¬P
    • ¸É§U³æ¦ì: ·L¬P¬ì§Þ
    • ­pµe°õ¦æ´Á¶¡: 2006/7/1 to 2007/6/30
    • ÃöÁäµü: »y­µ¦X¦¨¡B´O¤J¦¡¨t²Î¤§À³¥Î
    • ºK­n²¤¶:
      §¹¦¨PC/PDA¥­¥x¤§»y­µ¦X¦¨ºtºâªk¤Î®i¥Üµ{¦¡¡A¥Î¨Ó°µ¬°¾÷¾¹¤HÀ³µª¨t²Îªº°ò¦¡C

  69. ±j°·©ÊÃöÁäµü»y­µ¿ëÃÑ

    • ­^¤å¦WºÙ: Robust Keyword Spotting
    • ­pµe½s¸¹:
    • ¥D«ù¤H: ±i´¼¬P
    • ¸É§U³æ¦ì: ·L¬P¬ì§Þ
    • ­pµe°õ¦æ´Á¶¡: 2006/7/1 to 2007/6/30
    • ÃöÁäµü: ±j°·©Ê»y­µ¿ëÃÑ¡BÃöÁäµü»y­µ¿ëÃÑ
    • ºK­n²¤¶:
      §¹¦¨PC¥­¥x¤§±j°·©Ê»y­µÃöÁäµü¿ëÃѺtºâªk¤Î®i¥Üµ{¦¡¡A¥Î¨Ó°µ¬°¾÷¾¹¤Hťı¨t²Îªº°ò¦¡C

  70. ¤â«ù¦¡¸Ë¸mªº¤f»¡¤¤¤å»²§U¾Ç²ß§Þ³N

    • ­^¤å¦WºÙ: Computer-assisted Spoken Chinese Learning Systems for Hand-held Device
    • ­pµe½s¸¹:
    • ¥D«ù¤H: ±i´¼¬P
    • ¸É§U³æ¦ì: ¸êµ¦·|
    • ­pµe°õ¦æ´Á¶¡: 2006/3/1 to 2006/12/31
    • ÃöÁäµü: ¤¤¤å»y­µ¿ëÃÑ¡B¹q¸£»²§Uµo­µ°V½m¡B¹q¸£»²§U¤f»¡»y­µ¾Ç²ß
    • ºK­n²¤¶:
      ¥»­pµe±N±Ä¥Î¤¤¤å»y­µ¿ëÃÑ»P»y­µµû¤Àªº§Þ³N¡A¶}µo¤@®M¹q¸£»²§U¤f»¡¤¤¤åªº¾Ç²ß¨t²Î¡A¥HPDA/SmartPhoneµ¥¤â«ù¦¡¸Ë¸m¬°¹ê²{¥­¥x¡C¨Ï¥ÎªÌ¥u­n¹ï³Á§J­·Á¿¤@¥y¤¤¤å¡A¦¹¨t²Î§Y¥i®Ú¾Ú¿é¤J»y­µ¸ê®Æ¡A§ì¨ú­µ¦â¡]»y­µªº¤º®e¡^¡BÁn½Õ¡]°ê»y¥|­ÓÁn½Õªº¤ÀÃþ¡^¡BÃý«ß¡]¨C¤@­Ó­µªº®É¶¡ªøµu¡^¡B­µ¶q¡]¨C¤@­Ó­µªº±j«×¡^µ¥»y­µ¯S¼x¡AµM«á¶i¦æ©M¥Ø¼Ð»y¥y©Î¬O»y­µ¼Ò«¬ªº¤ñ¹ï¡A®Ú¾Ú¨ä¬Û¦ü«×¨Ó²£¥Í¤@­Ó¤¶©ó0©M100¤§¶¡ªº¤À¼Æ¡A¨Ã¥i¶i¦æ¿ù»~¤ÀªR¡Aµ¹¤©¨Ï¥ÎªÌµo­µ§ï¶iªº«Øij¡CÁöµM»y­µ¿ëÃѪº¨t²Î¤w¦³³\¦h¡A¦ý¬O¦h¥bÁÙ¬O¦bPC¥­¥xªºÀ³¥Î¡A¦]¦¹¥»­pµeªº³Ì¤j¬D¾Ô¡A´N¬O±N»y­µ¿ëÃÑ»Pµû¤À¡A²¾´Ó¨ì­pºâ¯à¤O¸û®tªº¤â«ù¦¡¸Ë¸m¡A³o¤è­±ªº§Þ³NªùÂe«Ü°ª¡A¥Ø«e¥«­±¤WÁÙ¨S¦³¬Ý¨ìÃþ¦üªº²£«~¡C

  71. ¥xÆW¼Æ¦ì¦³Án®Ñºô¤§«Ø¸m»P±À¼s

    • ­^¤å¦WºÙ: The Development and Promotion of the Web Portal for Digital Talking Books in Taiwan
    • ­pµe½s¸¹: NSC 95-2422-H-007 -001
    • ¥D«ù¤H: ±i´¼¬P¡B­ð¶Ç¸q¡B³¯©yªY
    • ¸É§U³æ¦ì: °ê¬ì·|
    • ­pµe°õ¦æ´Á¶¡: 2006/3/1 to 2007/2/28
    • ÃöÁäµü: DAISY Book¡B ¼Æ¦ì¦³Án®Ñ¡B»y­µ¿ëÃÑ¡B»y­µ¦X¦¨¡B»y­µÀ˯Á
    • ºK­n²¤¶:
      ¦b°ê»Ú¤WDAISY (Digital Accessible Information System) ¦³Án®Ñ¦bµø»Ù»P¾Ç»Ùªº¾Ç²ß¤è­±¡A¤w¸g¦¨¬°¤£¥i©Î¯Êªº¤u¨ã¡A¬°¤FÅý¥xÆWµø»ÙŪªÌ³£¥i¨É¦³¼Æ¦ì¾Ç²ßªºÀuÂI¡A¨Ã¥R¤À¨É¥ÎDAISY ¦³Án®Ñªº¯S¦â¡A¦]¦¹µo®i¤@®M¾A¦XµØ¤H¨Ï¥ÎªºÀô¹Ò¨è¤£®e½w¡C¦b93¡B94 ¦~ªº¼Æ¦ì¨åÂípµe¤¤¡A§Ú­Ì¤w¸g§¹¦¨¤F¨â¶µ¬ÛÃö¨t²Î¡G
      1. ¥xÆWµø»Ù¥Î¼Æ¦ì¨åÂ䧻y­µÀ˯Á¨t²Î¡G¦¹¨t²Î¥iÅý¨Ï¥ÎªÌª½±µ¸g¥ÑÁn±±ªº¤è¦¡¡AÀ˯Á¦³Án®Ñ¹Ï®ÑÀ]¤ºªºÂîѡA¨Ã¥i¥H¨Ï¥Î»y­µ¦X¦¨ªº¤è¦¡¡A¨Ó°á¥X®Ñ¦W¡A¥H«K¨Ï¥ÎªÌ¤U¸ü²âÅ¥¡C¡]¦¹¨t²Î¤w¸g¹ê»Ú¤W½u¡A¨Ã¥Î©ó²M¤jª¼¤Í·|¡A¥Î¤á¤w¸g¹F¨ì283 ¤H¡A¤U¸ü¦¸¼Æ²Ö­p7326¦¸¡A94/11/28 ¤î¡^¡C
      2. ¨ã¦³Án±±¥\¯àªºDAISY ¤¤¤å¼½©ñ¾¹¡G³o¬O¤@­Ó¥HJava µ{¦¡»y­µ©Ò½s¼gªºDAISY ¤¤¤å¼½©ñ¾¹¡A¾A¥Î©ó¤@¯ëPC¡A¦P®É¥[¤W¤F»y­µÀ˯Á»P»y­µ¦X¦¨ªº¥\¯à¡A¬O¥þ²y²Ä¤@­Ó¤¤¤åDAISY¼½©ñ¾¹¡A¤]¬O¥þ²y¥Ø«e°ß¤@¨ã¦³»y­µÀ˯Á»P»y­µ¦X¦¨¥\¯àªºDAISY ¼½©ñ¾¹¡C
      °ò©ó«e­z¨â¶µ¨t²Îªº¦¨¥\¡A§Ú­Ì§Æ±æ¦b95 ¦~«×ªº­pµe¤¤¡A»s§@¤@­ÓDAISY ¦³Án®Ñºô¡A¨Ã±NDAISY ¤¤¤å¼½©ñ¾¹²¾´Ó¨ì¦UºØ¥i¥Îªº¥­¥x¡]¥]§tPC¡BPDA¡BSmartPhone¡^¡A±À¼sµ¹µø»Ù»P¾Ç»Ù¨Ï¥Î¡A¨Ã¬ã¨s¸Ñ¨M´¼°]Åvªº°ÝÃD¡A±´°Q±À¼s¦Ü¤@¯ë¥Á²³¡A´Á±æ¹F¦¨ªº¨ãÅé®ÄªG¦³¡G
      1. ´£¨Ñµø»ÙªÌ¤@­Ó«K±¶¨ú±o¦³Án®ÑªººÞ¹D¡A¥H­°§C¼Æ¦ì¸¨®t¡C
      2. ´£¨Ñ¥i©ó¤£¦P¥­¥x¡]¥]§tPC¡BPDA¡BSmartPhone¡^¨Ï¥ÎªºDAISY ¤¤¤å¼½©ñ¾¹¡A¨Ã¨ã¦³»y­µÀ˯Á»P¦X¦¨ªº¥\¯à¡C
      3. §Q¥Îºô¸ô¬ì§Þ¾ã¦X°ê¤º¦U¦³Án»s§@³æ¦ì¡A¥H¸`¬Ù¾ãÊ^ªÀ·|¦¨¥»¡C
      4. §Q¥Î¼Ð·Ç¤Æªº¼Æ¦ì¦³Án®Ñ»s§@³W½d¡A¤è«K¬y³q³æ¦ì¶¡ªº¥æ¬y¡C
      5. ¤Þ¶i°ê»Ú¦³Án®Ñ»s§@§Þ³N¡A¤è«KŪªÌťŪ¡A¥H¼W¶i¦³Án®Ñªº¬y³q¡C
      6. §Q¥Î¼Æ¦ì»s§@¡B¨åÂûP¬y³q¡A²Å¦X°ê»ÚÁͶաA¬°°ê»Ú¦³Án®Ñ¥æ¬yçE¸ô¡C
      7. °£¤F¯à°÷¨ã¦³»y­µÀ˯Á¥\¯à¥~¡A¤]¯à°÷´£¨Ñ¦ÛµM®®¦Õªº»y­µ¦X¦¨®ÄªG¡A¥H­°§C¦³Án®Ñªº»s§@¦¨¥»¡C

  72. ­µ¼Ö·j´Mªº¥[³t»P¿ëÃѲv´£¤É¡A¤Î¨ä¦b´O¤J¦¡¨t²Îªº¹ê§@»PÀ³¥Î¡]2/3¡^

    • ­^¤å¦WºÙ: Speedup Mechansim and Performance Enhancement for Music Information Retrieval, with Applications to Embedded Systems
    • ­pµe½s¸¹: NSC 93-2213-E-007-058
    • ¥D«ù¤H: ±i´¼¬P
    • ¸É§U³æ¦ì: °ê¬ì·|
    • ­pµe°õ¦æ´Á¶¡: 2005/8/1 to 2006/7/31
    • ÃöÁäµü: ­µ¼Ö¸ê°TÀ˯Á¡B¦h¼Ò¦¡¨Ï¥ÎªÌ¤¶­±¡B±Û«ß¿ëÃÑ¡B­µ°ª°lÂÜ¡B¦h´CÅé¸ê®Æ®wÀ˯Á¡B°ÊºA®É¶¡®Õ¥¿
    • ºK­n²¤¶:
      ¥Ñ©ó¥»­pµe¥D«ù¤H¹L¥h¼Æ¦~¦b°ê¬ì·|¡B±Ð¨|³¡¨ô¶V­pµe¤Î²M½«¬ì§Þ¤§¸g¶O¤ä´©¤U¡A¤w¸g¦¨¥\¦a«Ø¥ß¤F¤@­Ó¤j«¬ªº­µ¼ÖÀ˯Á¨t²Î¡A¦¬¶°¤F¶W¹L¨â¸U­ººq¦±¡AÅý¨Ï¥ÎªÌ¯à¥H­ó°Û¡B»y­µ»PºVÀ»µ¥¤£¦P¤è¦¡¨Ó¿ï¨úºô»Úºô¸ô¤Wªººq¦±¡C¬°¤F¯à°÷¥H§Ö³tªº¤èªk¹ï¤j¶q¸ê®Æ¶i¦æ¦³®Äªº¤ñ¹ï¡A§Ú­Ì°£¤F¦b¤ñ¹ï¤èªk¶i¦æ²`¤J¬ã¨s¥~¡A¤]¬[³]¤Fºô®æ¹Bºâ¡]Grid Computing¡^ªº¦øªA¾¹¥­¥x¡A´Á±æ¥H¥­¦æ¤Î¤À´²³B²zªº¤è¦¡¡A¼W¶i¨t²Îªº¥i¥Î©Ê¡C¦b¦¹ºØ¥]§t¤j¶q¸ê®Æ³B²z»P¤ñ¹ïªº¨t²Î¤¤¡A­l¥Í¤F³\¦h¬ÛÃö¬ã¨sªº¬ã¨sijÃD¡A¥»¬ã¨s­pµeÀÀ¥H¤T¦~®É¶¡¡A°w¹ï³o¨ÇijÃD¶i¦æ²`¤J±´°Q¡A³o¨ÇijÃD¤Î¬ÛÃö¬ã¨s®Éµ{¦C¥X¦p¤U¡G
      1. ²Ä¤@¦~¡G ¥H´£°ª¿ëÃѲv¬°¥D
        1. ­µ°ª°lÂÜ¡]Pitch Tracking¡^ªº¦UºØ·s¤èªkªº¹Á¸Õ¡A¥]§t Combination of Classifiers¡BACF/AMDF Map µ¥¡C
        2. °ÊºA®É¶¡®Õ¥¿¡]Dynamic Time Warping¡^ªº§ï¶i»P´ú¸Õ¡A¥]§t Local Path Constraints¡BPath Penalty¡BFlexible Starting Points µ¥¡C
      2. ²Ä¤G¦~¡G¥H¥[§Ö¿ëÃѳt«×¬°¥D
        1. ¡u¨â¶¥¬q¦¡¤ñ¹ïªk¡vªº±´°Q»P§ï¨}¡A¥]§t¥H¤£¦Pªº¤è¦¡¶i¦æ¨â¶¥¬q¤ñ¹ï¡A¨Ò¦p Linear Scaling ¥[ DTW (Dynamic Time Warping)¡B
        2. ¨ä¥L¦UºØ­µ¼ÖÀ˯Á¤èªkªº±´°Q»P¥[³t¡A¥]§t¦p¦óºë½T¦a­pºâ DTW ªº Lower Bound¡B¦p¦ó¥H FastMap ªº¤èªk¨Ó±Nºq¦±Âà´«¨ì°ª¬°ªÅ¶¡¤¤ªºÂIµ¥¡C
        3. ÁôÂæ¡°¨¥i¤Ò¼Ò«¬¡]Hidden Markov Models¡^ªº¹Á¸Õ»P§ï¶i
      3. ²Ä¤T¦~¡G¥H´O¤J¦¡¨t²Îªº²¾´Ó¬°¥D
        1. ­µ°ª°lÂÜ»PÀ˯Á¤èªkªººë¶i»P²¤Æ
        2. ´O¤J¦¡¨t²Îªº¹ê§@»PÀ³¥Î

  73. §Q¥Î¥xÆW²{¦³µø»Ù¥Î¼Æ¦ì¨åÂøê®Æ»s§@ª¼¤H¹q¤l®Ñ-µo®iDAISY¤¤¤å¼½©ñ¾¹¤Î¬ÛÃö¤§»y­µ¿ëÃÑ»P¦X¦¨¥\¯à(¤l­pµe¤G)

    • ­^¤å¦WºÙ: On the Development of DAISY Chinese Player with Speech-enabled Interface Based on Speech Recognition and Synthesis
    • ­pµe½s¸¹: NSC 94-2422-H-007-005
    • ¥D«ù¤H: ±i´¼¬P
    • ¸É§U³æ¦ì: °ê¬ì·|
    • ­pµe°õ¦æ´Á¶¡: 2005/3/1 to 2006/2/28
    • ÃöÁäµü: DAISY, speech recognition, speech synthesis
    • ºK­n²¤¶:
      ¥»¬ã¨s­pµeªº¥D­n¥Ø¼Ð¡A¦b«Ø¥ß¤@­Ó¤¤¤å¤Æªº DAISY ³nÅé¼½©ñ¾¹¡A¥HªA°È¥xÆW¦a°Ïªºµø»Ù¦P­M¡A¨Ã¦b¦¹³nÅé¥[¤W»y­µ¿ëÃÑ»P¦X¦¨¥\¯à¡A¥H©µ¦ù­ì¥ý DAISY ³nÅé¼½©ñ¾¹ªº¥\¯à¡A´£¨Ñª¼­M¥H»y­µ¨Ó¾Þ±±³nÅé©Î¶i¦æ¥þ¤åÀ˯Á¡A¨Ã±NÀ˯Á©Ò±oªº¸ê°T¥H»P­µªº¤è¦¡¨Ó§e²{¡C¥Ø«e´£¨Ñ¤½¶}¤U¸üªº DAISY ³nÅé¼½©ñ¾¹¤j·§¦³¤Q¦hºØ¡A¦ý¬O³£¬O°ê¥~µo®iªº¨t²Î¡A¨S¦³¤¤¤å¤Æªº¤¶­±¡A§Ú­Ì·|°Ñ¦Ò²{¦sªº¼½©ñ¾¹¡Aµo®i¥X¤¤¤å¤Æªº¨t²Î¡A¥H²Å¦X¥xÆW¦a°Ïªº»Ý¨D¡C¥t¥~¡A§Ú­Ì¤]·|±N»y­µ¦X¦¨»P»y­µ¿ëÃÑ¥[¤J¦¹¨t²Î¡AÅý¨Ï¥ÎªÌ¥i¥H¨Ï¥Î»y­µ©R¥O¨Ó·j´M DAISY ¤º®e¡A­Y¥u§ä¨ì¤å¦r¤º®e¡A¨t²Î¤]¥i¥H¨Ï¥Î»y­µ¦X¦¨¨Ó¼½©ñ¡A©µ¦ù­ì¥ýªº DAISY ¼½©ñ¾¹ªº¥\¯à¡AÅýª¼­M¤Îµø»ÙªB¤Í§ó©ö©ó¨Ï¥Î¦¹¨t²Î¡A­°§C¦]¬°¨­Åé»Ùê©Ò¥i¯à³y¦¨ªº¼Æ¦ì¸¨®t¡C

  74. ¦h¼Ò¦¡­µ¼ÖÀ˯Áªº¥[³t¤èªk

    • ­^¤å¦WºÙ: Speedup Mechanisms for Multi-modal Music Information Retrieval
    • ­pµe½s¸¹: NSC 93-2622-E-007-012-CC3
    • ¥D«ù¤H: ±i´¼¬P
    • ¸É§U³æ¦ì: °ê¬ì·|¡B²M½«¬ì§Þ
    • ­pµe°õ¦æ´Á¶¡: 2004/11/1 to 2005/10/31
    • ÃöÁäµü: ¦h¼Ò¦¡­µ¼Ö¸ê°TÀ˯Á¡B¹Ï§Î¿ëÃÑ¡B°ÊºA³W¹º¡B°òÀW¤ÀªR¡B¼Æ¦ì°T¸¹³B²z
    • ºK­n²¤¶:
      ºô¸ô©Ò¼Ðº]ªº²×·¥¥Ø¼Ð¡A¬O­n¨Ï¥ô¦ó¤H¯à¦b¥ô¦ó®É¶¡¡B¥ô¦ó¦aÂI¡A³£¯àÀò±o©Ò»Ýªº¥ô¦ó¸ê°T©ÎªA°È¡A¦]¦¹´£¨Ñ¨Ï¥ÎªÌ¤@­Ó¦³®Äªº¸ê°TÀ˯Á¾÷¨î¡A«K¦¨¬°À˯Á¨t²Î©ÒÀ³¸Ó¦Ò¼{ªº­«­n½ÒÃD¡C¥»­pµeªº°Ê¾÷§Y¬O§Æ±æ°w¹ï­µ¼ÖÀ˯Á¨t²Î¡A´£¨Ñ¦h¼Ò¦¡ªº­µ¼ÖÀ˯Á§Þ³N°t¦X­µ¼ÖÀ˯Áªº¤Íµ½¤¶­±¡A¨Ãµo®i¦³®Äªº¥[³t¤èªk¡A¥H­°§C­pºâ®É¶¡¡A¨Ã¨Ï¤ñ¹ï§Þ³N¯à¯u¥¿¸¨¹ê©ó¹ê»Ú¥i¥Îªº°Ó«~¤W¡C¦b¦h¼Ò¦¡À˯Á§Þ³N¤W¡A§Ú­Ì´£¨Ñquery by singing¡Bquery by humming¡Bquery by tapping ¥H¤Î query by speech¡AÅý¨Ï¥ÎªÌ¯à¦h¤¸¤Æ¿é¤JÀ˯Á¸ê®Æ¡C¦b¥[³t¬ã¨s¤èªk¤W¡A§Ú­Ì·|±Ä¥Î«Ø¥ß¯Á¤Þ(indexing)¬°°ò¦ªº¤èªk¡A¨Ó­°§C¹ê»Ú­pºâªº®É¶¡¡C¤]´N¬Oµ²¦X¶¥¼h¦¡ªº¤ñ¹ï¤è¦¡¡]¥ý¥Î§Ö³t¤èªk§R°£¤£¥i¯àªººq¡A¦A¥Î¸Ô²Óªº¤ñ¹ï¤èªk¨Ó³B²z³Ñ¾lªººq¡^¡A¨Ã§Q¥Î¼Æ¾Ç¤ÀªR§ä¥X¤ñ¹ï¤èªk¤¤ªº³Ì¨Î°Ñ¼Æ­È¡C³Ì«á¡A§Ú­Ì­n§Q¥Î¤j¶qªº¿ý­µ¸ê®Æ¨Ó¹ï¨t²Î®Ä¯à¶i¦æ³Ì¨Î¤Æ¡A¥H¨D¨ú³Ì¨Îªº¨t²Î°Ñ¼Æ¡C¥»­pµeªº³Ì²×¥Ø¼Ð¡A´N¬O§Æ±æ¨Ï¥ÎªÌ¯àª½±µ¹ï³Á§J­·­ó°Û¤K¬í¡A§Q¥Î¥»­pµe©Ò¶}µoªº­µ¼ÖÀ˯Á¨t²Î§Y¥i¹ï¤j«¬ºq¦±¸ê®Æ®w¡]¥]§t¤T¸U­ººq¦±¡^¶i¦æ¦³®Ä¥B¨³³tªº¤ñ¹ï¡A¨Ã¯à¦b¤­¬í¡]±qÀY¤ñ¹ï¡^©Î¤Q¤­¬í¡]±q¥ô·N³B¤ñ¹ï¡^¤º¦^¶Ç¤ñ¹ï³Ì¬Û¦üªº«e¤G¤Q­ººq¡C

  75. ¥xÆWµø»Ù¥Î¼Æ¦ì¨åÂ䧻y­µÀ˯Á¨t²Î

    • ­^¤å¦WºÙ: Speech-based Information Retrieval for the Blind by Digital Archives in Taiwan
    • ­pµe½s¸¹:
    • ¥D«ù¤H: ­ð¶Ç¸q¡B±i´¼¬P
    • ¸É§U³æ¦ì: °ê¬ì·|
    • ­pµe°õ¦æ´Á¶¡: 2004/8/1 to 2005/4/30
    • ÃöÁäµü: ¼Æ¦ì°T¸¹³B²z¡B»y­µ¿ëÃÑ¡B¹Ï§Î¿ëÃÑ
    • ºK­n²¤¶:
      ·í§Ú­Ì¹ï¶g¾D¨Æª«µø­YµL¸@®É¡A¦³¨Ç¤H¡B«o¥u¯à³z¹LºN¯Á¨D¨ú¥Í¦s¡C·íºô¸ô¥[³t¸ê°Tªº¬y³q®É¡A¦³¨Ç¤H¡B«o³vº¥³à¥¢¨Dª¾ªº°ò¥»¤HÅv¡C¬O§_¼Æ¦ì¨åÂêºÀ³¥Î¥[­È§Þ³N¯à§ïµ½¦¹ª¾ÃѸ¨®t(Knowledge Gap)¡H

      ¥»¬ã¨s¥Øªº¬°«Ø¥ß¥xÆWµø»Ù¥Î¼Æ¦ì¨åÂ䧻y­µÀ˯Á¨t²Î¡C¥»­pµe¥H±Nºô»Úºô¸ô·s»D¤Î­µ¼ÖÀ˯Á¬°¬ã¨sµo®iªº¥Ø¼Ð¡A¥D­n´£¨Ñ¤F¥þ¤è¦ìª¼¤H¡y®ÑÄy¡z¡B¡y·s»D¡z¡B¡y­µ¼Ö¡zªºÀ˯ÁªA°È¡A³z¹L¤H©Ê¤Æªº¨Ï¥ÎªÌ¤¶­±¡A¥H»y­µ¿é¤Jªº¤è¦¡¨Ó·j´M·s»D¤Î®ÑÄy¡A¥H­ó°Ûªº¤è¦¡¨Ó·j´M­µ¼Ö¡A¨Ã«Ø¥ß¤À´²¦¡¼Æ¦ì¦³Án¸ê®Æ®w¡AÅýª¼¤Í¯àÂǥѻy­µªºÀ˯Á¡A§ó¤è«K¡B¦³®Ä²vªº¦s¨ú¼Æ¦ì¸ê°T¡C

      ¦bª¼¤Í·|«Ø¸m¡yÁp¦X®Ñ¥Ø³q³ø¨t²Î¡z¡B¡y¥þ°êµø»Ù¦³Án®Ñ»s§@¥­¥x¡z¡B¡yª¼¤H¦³Án®Ñ¹Ï®ÑÀ]«Ø¸m­pµe¡z¤Î¦Û¶O±N¤j¶q»y­µ¦³Án®Ñ¼Æ¦ì¤Æ«á¡A³z¹L¥»­pµe¹ê§@¤§»y­µ¦³ÁnÀ˯Á¨t²Îªº¥[­ÈÀ³¥Î¡A±N¥i¥H§¹¾ã¨ó½Õ¾ã¦X°ê¤ºª¼¤Í¬ÛÃöªA°È¸ê·½¡A¨Ã¯u¥¿´£¤Éª¼¤ÍªA°È³æ¦ì¤§«e©Ò¶i¦æ¬Ã¶Q¼Æ¦ì¨åÂ䧪þ¥[»ù­È¡A¸É¨¬ª¼¤Í¼Æ¦ì¨åÂë÷¹Ï¤¤³Ì«á¤@¶ô¤]¬O³Ì­«­n¤@¶ôªº¯Ê¨¤¡C¬ÛÃöªº§Þ³N¤Î¨å½d±N¨Ó¨Ã¥i²¾Âà¦Ü°ê¤º¬ÛÃöªº¤å±ÐªA°È³æ¦ì¤Î¼Æ¦ì¨åÂþ÷ºc¡A±N¨Ó§ó¥i¹Á¸Õ¬ãÀÀ¥X¹ï¤@¯ëªB¤Íªº¦¬¶O¥[­ÈªA°È¡A¥H¹ï©ó¦U®z¶Õ¹ÎÅé¾÷ºc¼W¥[¥~¦¬¤J¡A«Ø¥ß¨ä¥ÃÄò¸gÀ礧¹B§@¾÷¨î¡A¦¨¬°°ê¤º¥~¼Æ¦ì¨åÂ䧤½¯q¥[­ÈÀ³¥Î¾ð¥ß¤@³Ì¨Î¨å½d¡C

  76. ­µ¼Ö·j´Mªº¥[³t»P¿ëÃѲv´£¤É¡A¤Î¨ä¦b´O¤J¦¡¨t²Îªº¹ê§@»PÀ³¥Î¡]1/3¡^

    • ­^¤å¦WºÙ: Speedup Mechansim and Performance Enhancement for Music Information Retrieval, with Applications to Embedded Systems
    • ­pµe½s¸¹: NSC 93-2213-E-007-058
    • ¥D«ù¤H: ±i´¼¬P
    • ¸É§U³æ¦ì: °ê¬ì·|
    • ­pµe°õ¦æ´Á¶¡: 2004/8/1 to 2005/7/31
    • ÃöÁäµü: ­µ¼Ö¸ê°TÀ˯Á¡B¦h¼Ò¦¡¨Ï¥ÎªÌ¤¶­±¡B±Û«ß¿ëÃÑ¡B­µ°ª°lÂÜ¡B¦h´CÅé¸ê®Æ®wÀ˯Á¡B°ÊºA®É¶¡®Õ¥¿
    • ºK­n²¤¶:
      ¥Ñ©ó¥»­pµe¥D«ù¤H¹L¥h¼Æ¦~¦b°ê¬ì·|¡B±Ð¨|³¡¨ô¶V­pµe¤Î²M½«¬ì§Þ¤§¸g¶O¤ä´©¤U¡A¤w¸g¦¨¥\¦a«Ø¥ß¤F¤@­Ó¤j«¬ªº­µ¼ÖÀ˯Á¨t²Î¡A¦¬¶°¤F¶W¹L¨â¸U­ººq¦±¡AÅý¨Ï¥ÎªÌ¯à¥H­ó°Û¡B»y­µ»PºVÀ»µ¥¤£¦P¤è¦¡¨Ó¿ï¨úºô»Úºô¸ô¤Wªººq¦±¡C¬°¤F¯à°÷¥H§Ö³tªº¤èªk¹ï¤j¶q¸ê®Æ¶i¦æ¦³®Äªº¤ñ¹ï¡A§Ú­Ì°£¤F¦b¤ñ¹ï¤èªk¶i¦æ²`¤J¬ã¨s¥~¡A¤]¬[³]¤Fºô®æ¹Bºâ¡]Grid Computing¡^ªº¦øªA¾¹¥­¥x¡A´Á±æ¥H¥­¦æ¤Î¤À´²³B²zªº¤è¦¡¡A¼W¶i¨t²Îªº¥i¥Î©Ê¡C¦b¦¹ºØ¥]§t¤j¶q¸ê®Æ³B²z»P¤ñ¹ïªº¨t²Î¤¤¡A­l¥Í¤F³\¦h¬ÛÃö¬ã¨sªº¬ã¨sijÃD¡A¥»¬ã¨s­pµeÀÀ¥H¤T¦~®É¶¡¡A°w¹ï³o¨ÇijÃD¶i¦æ²`¤J±´°Q¡A³o¨ÇijÃD¤Î¬ÛÃö¬ã¨s®Éµ{¦C¥X¦p¤U¡G
      1. ²Ä¤@¦~¡G ¥H´£°ª¿ëÃѲv¬°¥D
        1. ­µ°ª°lÂÜ¡]Pitch Tracking¡^ªº¦UºØ·s¤èªkªº¹Á¸Õ¡A¥]§t Combination of Classifiers¡BACF/AMDF Map µ¥¡C
        2. °ÊºA®É¶¡®Õ¥¿¡]Dynamic Time Warping¡^ªº§ï¶i»P´ú¸Õ¡A¥]§t Local Path Constraints¡BPath Penalty¡BFlexible Starting Points µ¥¡C
      2. ²Ä¤G¦~¡G¥H¥[§Ö¿ëÃѳt«×¬°¥D
        1. ¡u¨â¶¥¬q¦¡¤ñ¹ïªk¡vªº±´°Q»P§ï¨}¡A¥]§t¥H¤£¦Pªº¤è¦¡¶i¦æ¨â¶¥¬q¤ñ¹ï¡A¨Ò¦p Linear Scaling ¥[ DTW (Dynamic Time Warping)¡B
        2. ¨ä¥L¦UºØ­µ¼ÖÀ˯Á¤èªkªº±´°Q»P¥[³t¡A¥]§t¦p¦óºë½T¦a­pºâ DTW ªº Lower Bound¡B¦p¦ó¥H FastMap ªº¤èªk¨Ó±Nºq¦±Âà´«¨ì°ª¬°ªÅ¶¡¤¤ªºÂIµ¥¡C
        3. ÁôÂæ¡°¨¥i¤Ò¼Ò«¬¡]Hidden Markov Models¡^ªº¹Á¸Õ»P§ï¶i
      3. ²Ä¤T¦~¡G¥H´O¤J¦¡¨t²Îªº²¾´Ó¬°¥D
        1. ­µ°ª°lÂÜ»PÀ˯Á¤èªkªººë¶i»P²¤Æ
        2. ´O¤J¦¡¨t²Îªº¹ê§@»PÀ³¥Î

  77. À³¥Î©ó¼Æ¦ì±Ð§÷ªº»y­µÀ˯Á»P»yªÌ½T»{

    • ­^¤å¦WºÙ: On the Use of Speech-based Retrieval and Speaker Verification for Digital Courseware
    • ­pµe½s¸¹:
    • ¥D«ù¤H: ±i´¼¬P
    • ¸É§U³æ¦ì: ²Î«H¥ø·~ºÞ²zÅU°Ý
    • ­pµe°õ¦æ´Á¶¡: 2004/3/1 to 2004/10/31
    • ÃöÁäµü: ¼Æ¦ì°T¸¹³B²z¡B»y­µ¿ëÃÑ¡B»yªÌ¿ëÃÑ¡B¹Ï§Î¿ëÃÑ
    • ºK­n²¤¶:
      ¥»­pµe±N±´°Q¦p¦ó¨Ï¥Î»y­µ§Þ³N©ó¼Æ¦ì±Ð§÷ªºÀ˯Á»P¦w¥þ¾÷¨î¡C¦bÀ˯Á¤è­±¡A§Ú­Ì±N¨Ï¥Î»y­µÀ˯Áªº¤è¦¡¡AÅý¨Ï¥ÎªÌ¥u­n¹ï³Á§J­·»¡¥XÃöÁäµü¡A¹q¸£§Y¥i·j´M¥X¬ÛÃöªº±Ð§÷¡C¦b¦w¥þ¾÷¨î¤è­±¡A§Ú­Ì¨Ï¥Î»yªÌ½T»{ªº§Þ³N¡AÅý¨Ï¥ÎªÌ¥u­n¹ïµÛ³Á§J­·»¡¥X¦Û¤vªº¥Nªí»y¥y¡A¨t²Î§Y¥i½T»{¦¹¨Ï¥ÎªÌ¬O§_¯u¬°¨ä¥»¤H¡C

  78. ¤j«¬­µ¼ÖÀ˯Á¨t²Îªº²z½×»P¹ê§@¡]3/3¡^

    • ­^¤å¦WºÙ: Large-scale Music Information Retrieval System: Theory and Implementation¡]3/3¡^
    • ­pµe½s¸¹: NSC 90-2213-E-007-058
    • ¥D«ù¤H: ±i´¼¬P
    • ¸É§U³æ¦ì: °ê¬ì·|
    • ­pµe°õ¦æ´Á¶¡: 2003/8/1 to 2004/7/31
    • ÃöÁäµü: ¦h´CÅé¸ê°TÀ˯Á¡B­µ¼Ö¸ê°TÀ˯Á¡B¹Ï§Î¿ëÃÑ¡BÃþ¯«¸gºô¸ô¡B°ÊºA³W¹º¡B°òÀW¤ÀªR¡B¼Æ¦ì°T¸¹³B²z
    • ºK­n²¤¶:
      ¥Ñ©óºô»Úºô¸ôªº¿³°_¡A¦UºØ¼Æ¦ì´CÅé¸ê°T¦bºô¸ô¤W¬y³qªº³t«×¶V¨Ó¶VÀWÁc¡A¼Æ¶q¤]¶V¨Ó¶V¤j¡A¦]¦¹¦p¦ó¯à¥H¦h´CÅé¸ê®Æªº¤º®e¡A¨Ó¹ï¦h´CÅé¸ê®Æ¶i¦æ¦³®Äªº¯Á¤Þ¡]Indexing¡^©MÀ˯Á¡]Retrieval¡A©Î½Ķ¦¨¡uÂ^¨ú¡v¡^¡A´NÅܦ¨¤@­Ó«D±`­«­nªº½ÒÃD¡A³o¤@Ãþªº¬ã¨s³q¦¨¬°¡u¥H¤º®e¬°¥Dªº¦h´CÅé¸ê°TÀ˯Á¡v¡]Content-based Multimedia Information Retrieval¡^¡A¦Ó¬ÛÃöªº¹ê§@¨t²Î¡A¤£¦ý¯à¥Î©óºô»Úºô¸ôªº¦h´CÅé·j´M¤ÞÀº¡A¤]¬O¼Æ¦ì¹Ï®ÑÀ]¡]©Î³Õª«À]¡^¤º¤£¥i©Î¯Êªº¦Û°Ê¤Æ¤u¨ã¡C

      ¥»­pµe±N¹ï­µ¼Ö¸ê°TÀ˯Á¶i¦æ§Þ³N»P²z½×ªº±´°Q¡A¨Ã±N¹ê§@¤@­Ó¤j«¬ªººô»Úºô¸ô­µ¼Ö·j´M¤ÞÀº¡A´Á±æ¯à±NÀ³¥Î©Ò¦³¸ê°TÀ˯Á¡B¹Ï§Î¿ëÃÑ¡BÃþ¯«¸gºô¸ô¡B¼Æ¦ì°T¸¹³B²z¡B°ÊºA³W¹º¡B³Ìªñ¾F©~·j´Mªk¡B»EÃþ¤ÀªRªº¦U¶µ²z½×»P¤èªk¡A¨Ó¹ï¤j«¬­µ¼Ö¸ê®Æ®w¶i¦æ§Ö³t¥B¦³®ÄªºÀ˯Á»P¤ñ¹ï¡A¨Ã¯àÀ³¥ÎWebµ{¦¡³]­pªº§Þ³N¡A¥H¤Î Client/server ¤À´²¦¡­pºâ¡ASever clusters¥­¦æ­pºâªº§Þ¥©¡A¨Ó¼W¥[À˯Á³t«×¡A¨Ï±o¨Ï¥ÎªÌ¹ï¤j«¬­µ¼Ö¸ê®Æ®w¶i¦æÀ˯Á®É¡A¥u­n¹ïµÛ³Á§J­·°Û¡]©Î­ó¡^8¬íÄÁ¡A¦¹¨t²Î§Y¥i¦b10¬í¤º¶Ç¦^³Ì¬Û¦üªº«e20¦Wºq¦±¡C

      ­n¶i¦æ¥»¶µ¬ã¨s¡A©Ò»Ýªº­I´ºª¾ÃѸû¦h¡A°£¤F­n¹ï¤W­z¦UºØ§Þ³N»P¤èªkªº´x´¤»P¹ê°µ¥~¡AÁÙ¥²¶·ÁA¸Ñ­µ¼Öªº¯S©Ê¡A¥H¤Î¨C­Ó¤H¤£¦Pªº°Ûºq²ß©Ê¡]Âà­µ¡B§Ý­µ¡B¦@»ïµ¥¡^¦b¼Æ¦ì°T¸¹¤Î°òÀW¤ÀªR©Ò³y¦¨ªº®ÄªG¡C¥»­pµe¥D«ù¤H±q1997¦~§Y¶}©l°ê¬ì·| VOD ­pµe¤§´¼¼z«¬¿ïºq¤¶­±ªº¬ã¨s¡A¨Ã«ùÄò¬ÛÃö¬ã¨s¦Ü¤µ¡A¨ä¶¡¨Ã©M²M½«¬ì§Þ¤½¥q¦X§@¡A¶}µo¥X¡u¶W¯ÅÂIºq¤ý¡v¡]½Ð¨£©Òªþ¤§¸Õ¥Î¥úºÐ¡^¡A¥i¹ï3000­ººq¶i¦æ¤Î®ÉÀ˯Á¡A¦]¦¹¤w²Ö¿n¬Û·í¬ã¨s°Ê¶q¡A§Æ±æ¯à¥Ó½Ð¦h¦~´Á¬ã¨s­pµe¡A¤@Á|«Ø¥ß¥þ²yÀ˯Á¥\¯à³Ì§¹µ½¡B¿ëÃѲv³Ì°ªªººô»Úºô¸ô­µ¼Ö·j´M¤ÞÀº¡A¤@¦p www.google.com ¦b¤å¦r·j´M¤ÞÀº©Ò¦ûªºÀsÀY¦a¦ì¯ë¡C

      ¦b²Ä¤@¦~ªº­pµe¤¤¡A§Ú­Ì§Æ±æ¯à«Ø¥ß¤@­Ó¤¤«¬­µ¼ÖÀ˯Á¨t²Î¡A¯à¹ï8000­ººq¦±¶i¦æÀ˯Á¡A¨Ã¯à¦b¤Q¬í¤º±o¨ìµª®×¡C§Ú­Ì±N¹Á¸Õ¦UºØ¥[³tªº¤èªk¡A¨Ã½Õ¾A¦UºØ¤èªkªº°Ñ¼Æ¡]¦p¶¥¼hÂo°£ªk¡A¸Ô¨£«á­z¡^¡A¥H¦b¦³­­ªº®É¶¡¤º¡A±o¨ì³Ì¨Îªº¿ëÃѲv¡C¦¹¥~¡A§Ú­Ì­n¥[±j°òÀW¤ÀªRªº¥i¾a«×¡A¨Ï±o¦¹¨t²Î¹ï©ó¨Ï¥Î¤£¦Pµo­µ¤èªkªº¤H¡A³£¯à§ì¨ú°÷¦hªº¥¿½T°òÀW¸ê°T¡A¥H¥[±j¾ã­Ó¨t²Îªº±j°·©Ê¡C

      ¦b²Ä¤G¦~ªº­pµe¤¤¡A§Ú­Ì§Æ±æ«ùÄò«e¤@¦~«×ªº¬ã¨s¡AÄ~Äò¹Á¸Õ¦UºØ¥[³t¤èªk¡]¸Ô¨£«á­z¡^¡A¨Ã´Á±æ±N³o¨Ç¤èªkÀ³¥Î©ó¤@­Ó¤j«¬­µ¼ÖÀ˯Á¨t²Î¡A¯à¹ï15000­ººq¶i¦æÀ˯Á¡A¨Ã¯à¦b¤Q¬í¤º±o¨ìµ²ªG¡C¦P®É§Ú­Ì¤]±N«Ø¥ß web ¤Wªº·j´M¤ÞÀº¡A¥H¤Î­µ¼ÖÀɦ۰ʻ`¶°µ{¦¡¤Î¦Û°Ê«Ø¥ß¯Á¤Þµ{¦¡¡A¨Ã§Q¥Î client/server ¶i¦æ¤À´²¦¡­pºâ¡A¥H´î»´ server ªº­pºâ¶q¥H¤Îºô¸ô¸ê®Æ¬y¶q¡C

      ¦b²Ä¤T¦~ªº­pµe¤¤¡A§Ú­Ì¤´µM­nºë¯q¨Dºë¡A¹Á¸Õ¦UºØ¨³³t¦³®Äªº¤ñ¹ï¤èªk¡C¦¹¥~¡A·íºq¶V¨Ó¶V¦h®É¡A§Ú­Ì­n¦Ò¼{«Ø¥ß server cluster¡A¥H¥­¦æ­pºâªº¤è¦¡¡A¨Ó­°§C¨Ï¥ÎªÌªºµ¥«Ý®É¶¡¡C¦P®É§Ú­Ì¤]­n¹ï server ÂIºq­pºâªº¦UºØ¸ê®Æ¶i¦æ¤ÀªR¡A¥H§ä¥X³Ì¦nªº task partitioning ªº¤è¦¡¡A¥H¤Î³Ì¨Îªº¤ñ¹ï¦¸§Ç¡C

  79. ­µ¼ÖÀ˯Áªº¥[³t¤èªk

    • ­^¤å¦WºÙ: Methods for Efficient Music Retrieval
    • ­pµe½s¸¹:
    • ¥D«ù¤H: ±i´¼¬P
    • ¸É§U³æ¦ì: °ê¬ì·|¡B²M½«¬ì§Þ
    • ­pµe°õ¦æ´Á¶¡: 2003/6/1 to 2004/5/31
    • ÃöÁäµü: ¦h´CÅé¸ê°TÀ˯Á¡B­µ¼Ö¸ê°TÀ˯Á¡B¹Ï§Î¿ëÃÑ¡BÃþ¯«¸gºô¸ô¡B°ÊºA³W¹º¡B°òÀW¤ÀªR¡B¼Æ¦ì°T¸¹³B²z
    • ºK­n²¤¶:
      ÀHµÛ¼Æ¦ì¤º®e¡]§tÁn­µ¡B­µ¼Ö¡B¼v¹³¡Bµø°Tµ¥¦h´CÅé¸ê®Æ¡^¼Æ¶qªº¤j´T¦¨ªø¡A¹ï©ó¼Æ¦ì¤º®eªºÀ˯Á¤]´NÅܦ¨¤@­Ó­«­nªº½ÒÃD¡C¥»­pµeªº°Ê¾÷§Y¬O§Æ±æ°w¹ï­µ¼ÖÀ˯Á¨t²Î¡A¶}µo¥H­ó°Û¿ïºqªº¤Íµ½¤¶­±¡A¨Ãµo®i¦³®Äªº¥[³t¤èªk¡A¥H­°§C­pºâ®É¶¡¡A¨Ã¨Ï¦¹¤ñ¹ï§Þ³N¯à¯u¥¿¸¨¹ê©ó¹ê»Ú¥i¥Îªº°Ó«~¤W¡C¦b¬ã¨s¤èªk¤W¡A§Ú­Ì·|±Ä¥Î¦UºØ nearest neighbor search ªº¦UºØ¤èªk¡]¦p k-d tree, branch-and-bound tree search, vantage-point tree search¡^¡A¨Ó­°§C¹ê»Ú­pºâªº®É¶¡¡C¦¹¥~¡A§Ú­Ì¤]·|¹Á¸Õ¶¥¼h¦¡ªº¤ñ¹ï¤è¦¡¡]¥ý¥Î§Ö³t¤èªk§R°£¤£¥i¯àªººq¡A¦A¥Î¸Ô²Óªº¤ñ¹ï¤èªk¨Ó³B²z³Ñ¾lªººq¡^¡A¨Ã§Q¥Î¼Æ¾Ç¤ÀªR§ä¥X¤ñ¹ï¤èªk¤¤ªº³Ì¨Î°Ñ¼Æ­È¡C³Ì«á¡A§Ú­Ì­n§Q¥Î¤j¶qªº¿ý­µ¸ê®Æ¨Ó¹ï¨t²Î®Ä¯à¶i¦æ³Ì¨Î¤Æ¡A¥H¨D¨ú³Ì¨Îªº¨t²Î°Ñ¼Æ¡C¥»­pµeªº³Ì²×¥Ø¼Ð¡A´N¬O§Æ±æ¨Ï¥ÎªÌ¯àª½±µ¹ï³Á§J­·­ó°Û¤K¬í¡A§Q¥Î¥»­pµe©Ò¶}µoªº­µ¼ÖÀ˯Á¨t²Î§Y¥i¹ï¤j«¬ºq¦±¸ê®Æ®w¡]¥]§t¤T¸U­ººq¦±¡^¶i¦æ¦³®Ä¥B¨³³tªº¤ñ¹ï¡A¨Ã¯à¦b¤­¬í¡]±qÀY¤ñ¹ï¡^©Î¤Q¤­¬í¡]±q¥ô·N³B¤ñ¹ï¡^¤º¦^¶Ç¤ñ¹ï³Ì¬Û¦üªº«e¤G¤Q­ººq¡C

  80. ­µ°T§Þ³N¬ã¨s

    • ­^¤å¦WºÙ: Studies on Audio Technology
    • ­pµe½s¸¹:
    • ¥D«ù¤H: ¾G¤h±d¡B¿à­¸ò¼¡B±i´¼¬P¡BĬ¤åà±
    • ¸É§U³æ¦ì: ­â¶§¬ì§Þ
    • ­pµe°õ¦æ´Á¶¡: 2002/9/1 to 2003/8/31
    • ÃöÁäµü: MP3½s½X¾¹¡B­µ¦âÀÉ®×À£ÁY¡B»y­µÂà´«¾¹¡B3D­µ°T³B²z
    • ºK­n²¤¶:
      ¥»­pµeµ²¦X¥x¤j¡B²M¤j¡B¦¨¤j­µ°T³B²z§Þ³N¬ã¨s¦³Ãö¾ÇªÌ¡A»P­â¶§¬ì§Þ¦P¤¯°Ó°Q«á¡A¨M©w¥HMP3 ½s½X¾¹³]­p¡B­µ¦âÀÉ®×À£ÁY¡B»y­µÂà´«¾¹³]­p¡B3D­µ°T³B²zµ¥¥|¶µ§Þ³N¬°¦X§@¬ãµo¶µ¥Ø¡C¯÷¤À¶µ»¡©ú¦p¤U¡G

      MP3½s½X¾¹¥Ñ¥x¤j­t³d¡A¶}µo¦X©óMPEG¼Ð·Çªº½s½X¾¹¡A¨Ã¹Á¸Õ´£¤É½s½X¾¹³nÅé§Þ³N¤ô·Ç¡A¨D§ó°ª«~½èªº­µ¼Ö»P§ó¤j¤ñ¨ÒªºÀ£ÁY¡C

      ­µ¦âÀÉ®×À£ÁY¥Ñ¥x¤j­t³d¡A¶}µo§ó¦³®Ä²vªº­µ¦âªi§ÎªíÀx¦s¤èªk¡A§Æ±æ¯à¸`¬ÙÀx¦sªÅ¶¡¡F±N¨Ó¥i¥H¹B¥Î¦b¤@¨Ç¸û¤pªº°O¾Ð¤¸¥ó©Î¬OÀx¦sªÅ¶¡¦³­­ªº³]³Æ¤W¡A´Á«Ø¥ß¥Ñsunplus´£¨Ñ¤§¥À­µ©Ò¹ïÀ³¤§À£ÁY¸ê®Æ®w¡C

      ¥H¤W¤G¶µÄÝ­µ¼Ö°T¸¹³B²z§Þ³N¡A¨ä¦¨ªG¥iÀ³¥Î©ó¦U¦¡¹q¤l¬ì§Þ²£«~¡A¨Ò¦p¤â¾÷¹aÁn¡AMP3ÀH¨­Å¥¡A¿ý­µµ§µ¥¡C

      »y­µÂà´«¾¹¥Ñ²M¤j­t³d¡A¶}µoÃþ¦ü¡u¬_«n¾÷¡vªº¤p¸Ë¸m¡A¥i¥H§Y®ÉÂà´«»y­µ¡C¦¹ºØ§Þ³NÀ³¥Î¼h­±¬Æ¼s¡A¨Ò¦p½ì¨ýª±¨ã¡B¹q¼v©Î¹qµø¼@ªº°t­µ¡B¦b¹ïÁ¿¾÷©Î¹q¸Ü¹ï½Í¤¤°°¸Ë­Ó¤Hªº¨­¥÷¡B²¾°£»yªÌªº­Ó¤H»y­µ¯S©Ê¥H¼W¶i»y­µ¿ëÃѨt²Îªº¿ëÃѲvµ¥¡C

      3D­µ°T³B²z¥Ñ¦¨¤j­t³d¡A¶}µoµêÀÀ5.1Án¹D­µÅT§Þ³N¦b´¶³q¦Õ¾÷²£¥Í°ª¯Å­µÅTªº®ÄªG¡A¦b¹q¤l¹CÀ¸¤ÎµêÀÀ¹ê¹Ò¤]¦³«Ü¦hÀ³¥Î¡C

  81. ¤j«¬­µ¼ÖÀ˯Á¨t²Îªº²z½×»P¹ê§@¡]2/3¡^

    • ­^¤å¦WºÙ: Large-scale Music Information Retrieval System: Theory and Implementation¡]2/3¡^
    • ­pµe½s¸¹: NSC 90-2213-E-007-058
    • ¥D«ù¤H: ±i´¼¬P
    • ¸É§U³æ¦ì: °ê¬ì·|
    • ­pµe°õ¦æ´Á¶¡: 2002/8/1 to 2003/7/31
    • ÃöÁäµü: ¦h´CÅé¸ê°TÀ˯Á¡B­µ¼Ö¸ê°TÀ˯Á¡B¹Ï§Î¿ëÃÑ¡BÃþ¯«¸gºô¸ô¡B°ÊºA³W¹º¡B°òÀW¤ÀªR¡B¼Æ¦ì°T¸¹³B²z
    • ºK­n²¤¶:
      ¥Ñ©óºô»Úºô¸ôªº¿³°_¡A¦UºØ¼Æ¦ì´CÅé¸ê°T¦bºô¸ô¤W¬y³qªº³t«×¶V¨Ó¶VÀWÁc¡A¼Æ¶q¤]¶V¨Ó¶V¤j¡A¦]¦¹¦p¦ó¯à¥H¦h´CÅé¸ê®Æªº¤º®e¡A¨Ó¹ï¦h´CÅé¸ê®Æ¶i¦æ¦³®Äªº¯Á¤Þ¡]Indexing¡^©MÀ˯Á¡]Retrieval¡A©Î½Ķ¦¨¡uÂ^¨ú¡v¡^¡A´NÅܦ¨¤@­Ó«D±`­«­nªº½ÒÃD¡A³o¤@Ãþªº¬ã¨s³q¦¨¬°¡u¥H¤º®e¬°¥Dªº¦h´CÅé¸ê°TÀ˯Á¡v¡]Content-based Multimedia Information Retrieval¡^¡A¦Ó¬ÛÃöªº¹ê§@¨t²Î¡A¤£¦ý¯à¥Î©óºô»Úºô¸ôªº¦h´CÅé·j´M¤ÞÀº¡A¤]¬O¼Æ¦ì¹Ï®ÑÀ]¡]©Î³Õª«À]¡^¤º¤£¥i©Î¯Êªº¦Û°Ê¤Æ¤u¨ã¡C

      ¥»­pµe±N¹ï­µ¼Ö¸ê°TÀ˯Á¶i¦æ§Þ³N»P²z½×ªº±´°Q¡A¨Ã±N¹ê§@¤@­Ó¤j«¬ªººô»Úºô¸ô­µ¼Ö·j´M¤ÞÀº¡A´Á±æ¯à±NÀ³¥Î©Ò¦³¸ê°TÀ˯Á¡B¹Ï§Î¿ëÃÑ¡BÃþ¯«¸gºô¸ô¡B¼Æ¦ì°T¸¹³B²z¡B°ÊºA³W¹º¡B³Ìªñ¾F©~·j´Mªk¡B»EÃþ¤ÀªRªº¦U¶µ²z½×»P¤èªk¡A¨Ó¹ï¤j«¬­µ¼Ö¸ê®Æ®w¶i¦æ§Ö³t¥B¦³®ÄªºÀ˯Á»P¤ñ¹ï¡A¨Ã¯àÀ³¥ÎWebµ{¦¡³]­pªº§Þ³N¡A¥H¤Î Client/server ¤À´²¦¡­pºâ¡ASever clusters¥­¦æ­pºâªº§Þ¥©¡A¨Ó¼W¥[À˯Á³t«×¡A¨Ï±o¨Ï¥ÎªÌ¹ï¤j«¬­µ¼Ö¸ê®Æ®w¶i¦æÀ˯Á®É¡A¥u­n¹ïµÛ³Á§J­·°Û¡]©Î­ó¡^8¬íÄÁ¡A¦¹¨t²Î§Y¥i¦b10¬í¤º¶Ç¦^³Ì¬Û¦üªº«e20¦Wºq¦±¡C

      ­n¶i¦æ¥»¶µ¬ã¨s¡A©Ò»Ýªº­I´ºª¾ÃѸû¦h¡A°£¤F­n¹ï¤W­z¦UºØ§Þ³N»P¤èªkªº´x´¤»P¹ê°µ¥~¡AÁÙ¥²¶·ÁA¸Ñ­µ¼Öªº¯S©Ê¡A¥H¤Î¨C­Ó¤H¤£¦Pªº°Ûºq²ß©Ê¡]Âà­µ¡B§Ý­µ¡B¦@»ïµ¥¡^¦b¼Æ¦ì°T¸¹¤Î°òÀW¤ÀªR©Ò³y¦¨ªº®ÄªG¡C¥»­pµe¥D«ù¤H±q1997¦~§Y¶}©l°ê¬ì·| VOD ­pµe¤§´¼¼z«¬¿ïºq¤¶­±ªº¬ã¨s¡A¨Ã«ùÄò¬ÛÃö¬ã¨s¦Ü¤µ¡A¨ä¶¡¨Ã©M²M½«¬ì§Þ¤½¥q¦X§@¡A¶}µo¥X¡u¶W¯ÅÂIºq¤ý¡v¡]½Ð¨£©Òªþ¤§¸Õ¥Î¥úºÐ¡^¡A¥i¹ï3000­ººq¶i¦æ¤Î®ÉÀ˯Á¡A¦]¦¹¤w²Ö¿n¬Û·í¬ã¨s°Ê¶q¡A§Æ±æ¯à¥Ó½Ð¦h¦~´Á¬ã¨s­pµe¡A¤@Á|«Ø¥ß¥þ²yÀ˯Á¥\¯à³Ì§¹µ½¡B¿ëÃѲv³Ì°ªªººô»Úºô¸ô­µ¼Ö·j´M¤ÞÀº¡A¤@¦p www.google.com ¦b¤å¦r·j´M¤ÞÀº©Ò¦ûªºÀsÀY¦a¦ì¯ë¡C

      ¦b²Ä¤@¦~ªº­pµe¤¤¡A§Ú­Ì§Æ±æ¯à«Ø¥ß¤@­Ó¤¤«¬­µ¼ÖÀ˯Á¨t²Î¡A¯à¹ï8000­ººq¦±¶i¦æÀ˯Á¡A¨Ã¯à¦b¤Q¬í¤º±o¨ìµª®×¡C§Ú­Ì±N¹Á¸Õ¦UºØ¥[³tªº¤èªk¡A¨Ã½Õ¾A¦UºØ¤èªkªº°Ñ¼Æ¡]¦p¶¥¼hÂo°£ªk¡A¸Ô¨£«á­z¡^¡A¥H¦b¦³­­ªº®É¶¡¤º¡A±o¨ì³Ì¨Îªº¿ëÃѲv¡C¦¹¥~¡A§Ú­Ì­n¥[±j°òÀW¤ÀªRªº¥i¾a«×¡A¨Ï±o¦¹¨t²Î¹ï©ó¨Ï¥Î¤£¦Pµo­µ¤èªkªº¤H¡A³£¯à§ì¨ú°÷¦hªº¥¿½T°òÀW¸ê°T¡A¥H¥[±j¾ã­Ó¨t²Îªº±j°·©Ê¡C

      ¦b²Ä¤G¦~ªº­pµe¤¤¡A§Ú­Ì§Æ±æ«ùÄò«e¤@¦~«×ªº¬ã¨s¡AÄ~Äò¹Á¸Õ¦UºØ¥[³t¤èªk¡]¸Ô¨£«á­z¡^¡A¨Ã´Á±æ±N³o¨Ç¤èªkÀ³¥Î©ó¤@­Ó¤j«¬­µ¼ÖÀ˯Á¨t²Î¡A¯à¹ï15000­ººq¶i¦æÀ˯Á¡A¨Ã¯à¦b¤Q¬í¤º±o¨ìµ²ªG¡C¦P®É§Ú­Ì¤]±N«Ø¥ß web ¤Wªº·j´M¤ÞÀº¡A¥H¤Î­µ¼ÖÀɦ۰ʻ`¶°µ{¦¡¤Î¦Û°Ê«Ø¥ß¯Á¤Þµ{¦¡¡A¨Ã§Q¥Î client/server ¶i¦æ¤À´²¦¡­pºâ¡A¥H´î»´ server ªº­pºâ¶q¥H¤Îºô¸ô¸ê®Æ¬y¶q¡C

      ¦b²Ä¤T¦~ªº­pµe¤¤¡A§Ú­Ì¤´µM­nºë¯q¨Dºë¡A¹Á¸Õ¦UºØ¨³³t¦³®Äªº¤ñ¹ï¤èªk¡C¦¹¥~¡A·íºq¶V¨Ó¶V¦h®É¡A§Ú­Ì­n¦Ò¼{«Ø¥ß server cluster¡A¥H¥­¦æ­pºâªº¤è¦¡¡A¨Ó­°§C¨Ï¥ÎªÌªºµ¥«Ý®É¶¡¡C¦P®É§Ú­Ì¤]­n¹ï server ÂIºq­pºâªº¦UºØ¸ê®Æ¶i¦æ¤ÀªR¡A¥H§ä¥X³Ì¦nªº task partitioning ªº¤è¦¡¡A¥H¤Î³Ì¨Îªº¤ñ¹ï¦¸§Ç¡C

  82. ­µ¼ÖÀ˯Áªº¥[³t¤èªk

    • ­^¤å¦WºÙ: Methods for Efficient Music Retrieval
    • ­pµe½s¸¹:
    • ¥D«ù¤H: ±i´¼¬P
    • ¸É§U³æ¦ì: °ê¬ì·|¡B²M½«¬ì§Þ
    • ­pµe°õ¦æ´Á¶¡: 2002/6/1 to 2003/5/31
    • ÃöÁäµü: ¦h´CÅé¸ê°TÀ˯Á¡B­µ¼Ö¸ê°TÀ˯Á¡B¹Ï§Î¿ëÃÑ¡BÃþ¯«¸gºô¸ô¡B°ÊºA³W¹º¡B°òÀW¤ÀªR¡B¼Æ¦ì°T¸¹³B²z
    • ºK­n²¤¶:
      ÀHµÛ¼Æ¦ì¤º®e¡]§tÁn­µ¡B­µ¼Ö¡B¼v¹³¡Bµø°Tµ¥¦h´CÅé¸ê®Æ¡^¼Æ¶qªº¤j´T¦¨ªø¡A¹ï©ó¼Æ¦ì¤º®eªºÀ˯Á¤]´NÅܦ¨¤@­Ó­«­nªº½ÒÃD¡C¥»­pµeªº°Ê¾÷§Y¬O§Æ±æ°w¹ï­µ¼ÖÀ˯Á¨t²Î¡A¶}µo¥H­ó°Û¿ïºqªº¤Íµ½¤¶­±¡A¨Ãµo®i¦³®Äªº¥[³t¤èªk¡A¥H­°§C­pºâ®É¶¡¡A¨Ã¨Ï¦¹¤ñ¹ï§Þ³N¯à¯u¥¿¸¨¹ê©ó¹ê»Ú¥i¥Îªº°Ó«~¤W¡C¦b¬ã¨s¤èªk¤W¡A§Ú­Ì·|±Ä¥Î¦UºØ nearest neighbor search ªº¦UºØ¤èªk¡]¦p k-d tree, branch-and-bound tree search, vantage-point tree search¡^¡A¨Ó­°§C¹ê»Ú­pºâªº®É¶¡¡C¦¹¥~¡A§Ú­Ì¤]·|¹Á¸Õ¶¥¼h¦¡ªº¤ñ¹ï¤è¦¡¡]¥ý¥Î§Ö³t¤èªk§R°£¤£¥i¯àªººq¡A¦A¥Î¸Ô²Óªº¤ñ¹ï¤èªk¨Ó³B²z³Ñ¾lªººq¡^¡A¨Ã§Q¥Î¼Æ¾Ç¤ÀªR§ä¥X¤ñ¹ï¤èªk¤¤ªº³Ì¨Î°Ñ¼Æ­È¡C³Ì«á¡A§Ú­Ì­n§Q¥Î¤j¶qªº¿ý­µ¸ê®Æ¨Ó¹ï¨t²Î®Ä¯à¶i¦æ³Ì¨Î¤Æ¡A¥H¨D¨ú³Ì¨Îªº¨t²Î°Ñ¼Æ¡C¥»­pµeªº³Ì²×¥Ø¼Ð¡A´N¬O§Æ±æ¨Ï¥ÎªÌ¯àª½±µ¹ï³Á§J­·­ó°Û¤K¬í¡A§Q¥Î¥»­pµe©Ò¶}µoªº­µ¼ÖÀ˯Á¨t²Î§Y¥i¹ï¤j«¬ºq¦±¸ê®Æ®w¡]¥]§t¤T¸U­ººq¦±¡^¶i¦æ¦³®Ä¥B¨³³tªº¤ñ¹ï¡A¨Ã¯à¦b¤­¬í¡]±qÀY¤ñ¹ï¡^©Î¤Q¤­¬í¡]±q¥ô·N³B¤ñ¹ï¡^¤º¦^¶Ç¤ñ¹ï³Ì¬Û¦üªº«e¤G¤Q­ººq¡C

  83. »y­µ°T¸¹³B²z»P¿ëÃѪº³nµwÅé¹ê§@»P¾ã¦X

    • ­^¤å¦WºÙ: Audio Signal Processing and Recognition: Software/Hardware Implementation and Integration
    • ­pµe½s¸¹:
    • ¥D«ù¤H: ±i´¼¬P
    • ¸É§U³æ¦ì: Üg«ä¬ì§Þ
    • ­pµe°õ¦æ´Á¶¡: 2002/1/1 to 2002/12/31
    • ÃöÁäµü:
    • ºK­n²¤¶:
      ÀHµÛ­Ó¤H¹q¸£ªº³t«×¨C18­Ó¤ë½¤É¤@­¿¡A©Ò¦³¥H«e³Qµø¬°»Ý­n¤j¶q­pºâ¡B¥u¯à¦b¤u§@¯¸°õ¦æªºÀ³¥Î¡A²{¦b³£¥i¥HºCºC²¾Âà¨ì­Ó¤H¹q¸£°õ¦æ¡C¦]¦¹¡A»y­µ°T¸¹ªº³B²z»P¿ëÃÑ¡A«Kº¥º¥¦¨¬°«D±`¹ê¥Î©Êªº§Þ³N¡A¥i¥H¾ã¦X¨ì­Ó¤H¹q¸£ªº¦UºØ¬ÛÃöÀ³¥Î¡C¦¹¥~¡A§ó¥Ñ©ó¦UºØ¸ê°T®a¹q¡B¦æ°Ê³q°T¡]¦p PDA¡BSmartPhone¡^ªº§Ö³tµo®i¡A¥H»y­µ¬°¤¶­±ªº´¼¼z«¬¿é¤J¤è¦¡¤]¶V¨Ó¶V­«­n¡A³o¨Ç¹ê¥Îªº§Þ³N°£¤F­n¾a³nÅ骺µo®i¥~¡A§ó­«­nªº¬O¦p¦ó¦b«Üµuªº®É¶¡¤º¡A±N³nÅ骺µ{¦¡½XÅܦ¨µwÅ骺¤¸¥ó¡A¨Ã¯à¥ß¨è¾ã¦X¨ì¬ÛÃöªºÀ³¥Î¨t²Î¤§¤¤¡C

      ±q³nÅé¨ìµwÅ骺¹Lµ{¡A°£¤F­n±Nºtºâªk§ï¬°¾A¦XµwÅé¹Bºâªº¤èªk¥~¡AÁÙ­n¦Ò¼{¦bµwÅé¤è­±ªº¦UºØ¹ê»Ú­­¨î¡A¨Ò¦p°O¾ÐÅ骺­­¨î¡B¥Ñ¯BÂI¼ÆÂà¦Ü©wÂI¼Æ©Ò³y¦¨ªº­pºâ»~®t¡B­pºâ³t«×ªº­°§C¡B¿é¥X/¿é¤Jªº¤¶­±­­¨îµ¥¡A³o¨Ç³£¬O§Ú­Ì­n¦Ò¼{ªº¦]¯À¡C

      ¦b¶Ç²Îªº§@ªk¤¤¡A¥Ñ³nÅé¦ÜµwÅ骺Âà´«¹Lµ{§¹¥þ­n¾a¤H¤u¨Ó§¹¦¨¡A¨ä¤¤¥]§t¤F³\¦h¡u¹Á¸Õ¿ù»~¡vªº¹Lµ{¡A¥»­pµe±N¦¹¹Lµ{ºÉ¶q¦Û°Ê¤Æ¡A§Q¥Î MATLAB/Simulink/Toolbox ªº¦UºØ²{¦¨¥\¯à¡A¹Á¸Õ±N¦b Simulink Àô¹Ò¤U©Òµo®iªº¿ëÃѨt²Î¡]¦p»yªÌ¿ëÃÑ¡B»y­µ©R¥O¿ëÃÑ¡B¦±½Õ¿ëÃÑ¡^Âà´«¨ì TI ªº 67xx ¨t¦CªºµwÅ饭¥x¡C¦¹¥~¡A§Ú­Ì¤]±Nµo®i¬ÛÃöªºµo®iÀô¹Ò»P¤¶­±¡A¹Á¸Õ±N Simulink ªº¦UºØ¿ëÃѨt²ÎÂà´«¦Ü§ó§C¶¥ªºª©¤l¡A¦p 8051 µ¥¡C

  84. ¤j«¬­µ¼ÖÀ˯Á¨t²Îªº²z½×»P¹ê§@¡]1/3¡^

    • ­^¤å¦WºÙ: Large-scale Music Information Retrieval System: Theory and Implementation¡]1/3¡^
    • ­pµe½s¸¹: NSC 90-2213-E-007-058
    • ¥D«ù¤H: ±i´¼¬P
    • ¸É§U³æ¦ì: °ê¬ì·|
    • ­pµe°õ¦æ´Á¶¡: 2001/8/1 to 2002/7/31
    • ÃöÁäµü: ¦h´CÅé¸ê°TÀ˯Á¡B­µ¼Ö¸ê°TÀ˯Á¡B¹Ï§Î¿ëÃÑ¡BÃþ¯«¸gºô¸ô¡B°ÊºA³W¹º¡B°òÀW¤ÀªR¡B¼Æ¦ì°T¸¹³B²z
    • ºK­n²¤¶:
      ¥Ñ©óºô»Úºô¸ôªº¿³°_¡A¦UºØ¼Æ¦ì´CÅé¸ê°T¦bºô¸ô¤W¬y³qªº³t«×¶V¨Ó¶VÀWÁc¡A¼Æ¶q¤]¶V¨Ó¶V¤j¡A¦]¦¹¦p¦ó¯à¥H¦h´CÅé¸ê®Æªº¤º®e¡A¨Ó¹ï¦h´CÅé¸ê®Æ¶i¦æ¦³®Äªº¯Á¤Þ¡]Indexing¡^©MÀ˯Á¡]Retrieval¡A©Î½Ķ¦¨¡uÂ^¨ú¡v¡^¡A´NÅܦ¨¤@­Ó«D±`­«­nªº½ÒÃD¡A³o¤@Ãþªº¬ã¨s³q¦¨¬°¡u¥H¤º®e¬°¥Dªº¦h´CÅé¸ê°TÀ˯Á¡v¡]Content-based Multimedia Information Retrieval¡^¡A¦Ó¬ÛÃöªº¹ê§@¨t²Î¡A¤£¦ý¯à¥Î©óºô»Úºô¸ôªº¦h´CÅé·j´M¤ÞÀº¡A¤]¬O¼Æ¦ì¹Ï®ÑÀ]¡]©Î³Õª«À]¡^¤º¤£¥i©Î¯Êªº¦Û°Ê¤Æ¤u¨ã¡C

      ¥»­pµe±N¹ï­µ¼Ö¸ê°TÀ˯Á¶i¦æ§Þ³N»P²z½×ªº±´°Q¡A¨Ã±N¹ê§@¤@­Ó¤j«¬ªººô»Úºô¸ô­µ¼Ö·j´M¤ÞÀº¡A´Á±æ¯à±NÀ³¥Î©Ò¦³¸ê°TÀ˯Á¡B¹Ï§Î¿ëÃÑ¡BÃþ¯«¸gºô¸ô¡B¼Æ¦ì°T¸¹³B²z¡B°ÊºA³W¹º¡B³Ìªñ¾F©~·j´Mªk¡B»EÃþ¤ÀªRªº¦U¶µ²z½×»P¤èªk¡A¨Ó¹ï¤j«¬­µ¼Ö¸ê®Æ®w¶i¦æ§Ö³t¥B¦³®ÄªºÀ˯Á»P¤ñ¹ï¡A¨Ã¯àÀ³¥ÎWebµ{¦¡³]­pªº§Þ³N¡A¥H¤Î Client/server ¤À´²¦¡­pºâ¡ASever clusters¥­¦æ­pºâªº§Þ¥©¡A¨Ó¼W¥[À˯Á³t«×¡A¨Ï±o¨Ï¥ÎªÌ¹ï¤j«¬­µ¼Ö¸ê®Æ®w¶i¦æÀ˯Á®É¡A¥u­n¹ïµÛ³Á§J­·°Û¡]©Î­ó¡^8¬íÄÁ¡A¦¹¨t²Î§Y¥i¦b10¬í¤º¶Ç¦^³Ì¬Û¦üªº«e20¦Wºq¦±¡C

      ­n¶i¦æ¥»¶µ¬ã¨s¡A©Ò»Ýªº­I´ºª¾ÃѸû¦h¡A°£¤F­n¹ï¤W­z¦UºØ§Þ³N»P¤èªkªº´x´¤»P¹ê°µ¥~¡AÁÙ¥²¶·ÁA¸Ñ­µ¼Öªº¯S©Ê¡A¥H¤Î¨C­Ó¤H¤£¦Pªº°Ûºq²ß©Ê¡]Âà­µ¡B§Ý­µ¡B¦@»ïµ¥¡^¦b¼Æ¦ì°T¸¹¤Î°òÀW¤ÀªR©Ò³y¦¨ªº®ÄªG¡C¥»­pµe¥D«ù¤H±q1997¦~§Y¶}©l°ê¬ì·| VOD ­pµe¤§´¼¼z«¬¿ïºq¤¶­±ªº¬ã¨s¡A¨Ã«ùÄò¬ÛÃö¬ã¨s¦Ü¤µ¡A¨ä¶¡¨Ã©M²M½«¬ì§Þ¤½¥q¦X§@¡A¶}µo¥X¡u¶W¯ÅÂIºq¤ý¡v¡]½Ð¨£©Òªþ¤§¸Õ¥Î¥úºÐ¡^¡A¥i¹ï3000­ººq¶i¦æ¤Î®ÉÀ˯Á¡A¦]¦¹¤w²Ö¿n¬Û·í¬ã¨s°Ê¶q¡A§Æ±æ¯à¥Ó½Ð¦h¦~´Á¬ã¨s­pµe¡A¤@Á|«Ø¥ß¥þ²yÀ˯Á¥\¯à³Ì§¹µ½¡B¿ëÃѲv³Ì°ªªººô»Úºô¸ô­µ¼Ö·j´M¤ÞÀº¡A¤@¦p www.google.com ¦b¤å¦r·j´M¤ÞÀº©Ò¦ûªºÀsÀY¦a¦ì¯ë¡C

      ¦b²Ä¤@¦~ªº­pµe¤¤¡A§Ú­Ì§Æ±æ¯à«Ø¥ß¤@­Ó¤¤«¬­µ¼ÖÀ˯Á¨t²Î¡A¯à¹ï8000­ººq¦±¶i¦æÀ˯Á¡A¨Ã¯à¦b¤Q¬í¤º±o¨ìµª®×¡C§Ú­Ì±N¹Á¸Õ¦UºØ¥[³tªº¤èªk¡A¨Ã½Õ¾A¦UºØ¤èªkªº°Ñ¼Æ¡]¦p¶¥¼hÂo°£ªk¡A¸Ô¨£«á­z¡^¡A¥H¦b¦³­­ªº®É¶¡¤º¡A±o¨ì³Ì¨Îªº¿ëÃѲv¡C¦¹¥~¡A§Ú­Ì­n¥[±j°òÀW¤ÀªRªº¥i¾a«×¡A¨Ï±o¦¹¨t²Î¹ï©ó¨Ï¥Î¤£¦Pµo­µ¤èªkªº¤H¡A³£¯à§ì¨ú°÷¦hªº¥¿½T°òÀW¸ê°T¡A¥H¥[±j¾ã­Ó¨t²Îªº±j°·©Ê¡C

      ¦b²Ä¤G¦~ªº­pµe¤¤¡A§Ú­Ì§Æ±æ«ùÄò«e¤@¦~«×ªº¬ã¨s¡AÄ~Äò¹Á¸Õ¦UºØ¥[³t¤èªk¡]¸Ô¨£«á­z¡^¡A¨Ã´Á±æ±N³o¨Ç¤èªkÀ³¥Î©ó¤@­Ó¤j«¬­µ¼ÖÀ˯Á¨t²Î¡A¯à¹ï15000­ººq¶i¦æÀ˯Á¡A¨Ã¯à¦b¤Q¬í¤º±o¨ìµ²ªG¡C¦P®É§Ú­Ì¤]±N«Ø¥ß web ¤Wªº·j´M¤ÞÀº¡A¥H¤Î­µ¼ÖÀɦ۰ʻ`¶°µ{¦¡¤Î¦Û°Ê«Ø¥ß¯Á¤Þµ{¦¡¡A¨Ã§Q¥Î client/server ¶i¦æ¤À´²¦¡­pºâ¡A¥H´î»´ server ªº­pºâ¶q¥H¤Îºô¸ô¸ê®Æ¬y¶q¡C

      ¦b²Ä¤T¦~ªº­pµe¤¤¡A§Ú­Ì¤´µM­nºë¯q¨Dºë¡A¹Á¸Õ¦UºØ¨³³t¦³®Äªº¤ñ¹ï¤èªk¡C¦¹¥~¡A·íºq¶V¨Ó¶V¦h®É¡A§Ú­Ì­n¦Ò¼{«Ø¥ß server cluster¡A¥H¥­¦æ­pºâªº¤è¦¡¡A¨Ó­°§C¨Ï¥ÎªÌªºµ¥«Ý®É¶¡¡C¦P®É§Ú­Ì¤]­n¹ï server ÂIºq­pºâªº¦UºØ¸ê®Æ¶i¦æ¤ÀªR¡A¥H§ä¥X³Ì¦nªº task partitioning ªº¤è¦¡¡A¥H¤Î³Ì¨Îªº¤ñ¹ï¦¸§Ç¡C

  85. »y­µ»PºqÁn¦X¦¨

    • ­^¤å¦WºÙ: Speech and Singing Voice Synthesis
    • ­pµe½s¸¹:
    • ¥D«ù¤H: ±i´¼¬P¡B¶À²ÐµØ
    • ¸É§U³æ¦ì: ²M½«¬ì§Þ
    • ­pµe°õ¦æ´Á¶¡: 2001/8/1 to 2002/7/31
    • ÃöÁäµü: Audio signal processing, speech synthesis, singing voice synthesis, sinusoidal modeling
    • ºK­n²¤¶:
      ¥Ñ©ó¹q¸£»Pºô¸ô¬ì§Þªºµo¹F¡A¦h´CÅé¤H¾÷¤¶­±ªº»Ý¨D»PÀ³¥Î¼h¦¸¤]¶V¨Ó¶V¼sªx¡C¹ï©ó¦h´CÅ骺³B²z¡A¤£¥~¥G¿ëÃÑ¡B¦X¦¨¡BÀ£ÁY»P¶Ç°e¡C¦b¥ý«e¬ÛÃöªº­pµe¤¤¡A§Ú­Ì¤w¸g¶i¦æ¤F¹ï©ó»y­µ»P±Û«ßªº¿ëÃÑ¡A¦b¥»­pµe¤¤¡A§Ú­Ì±N¶i¤@¨B±´°Q»y­µ»PºqÁnªº¦X¦¨¡A¥H«Ø¥ß¹ê¥Î¤Æªº¦h´CÅé¤H¾÷¤¶­±¨t²Î¡C

  86. »y­µ»P­µ¼Ö°T¸¹ªº¤ñ¹ï¤èªk»P¥[³t¾÷¨î

    • ­^¤å¦WºÙ: Music/Speech Information Retrieval and Their Speedup Mechanisms
    • ­pµe½s¸¹:
    • ¥D«ù¤H: ±i´¼¬P
    • ¸É§U³æ¦ì: ²M½«¬ì§Þ
    • ­pµe°õ¦æ´Á¶¡: 2001/5/1 to 2002/7/31
    • ÃöÁäµü: ¦h´CÅé¸ê°TÀ˯Á¡B­µ¼Ö¸ê°TÀ˯Á¡B»y­µÀ˯Á¡B¹Ï§Î¿ëÃÑ¡B°ÊºA³W¹º¡B°òÀW¤ÀªR¡B¼Æ¦ì°T¸¹³B²z
    • ºK­n²¤¶:
      ÀHµÛ¹q¸£»Pºô¸ô ¬ì§ÞªºÃz¬µ©Êµo®i¡A¦h´CÅé¸ê°T¦bºô¸ô¬y³qªº¼Æ¶q¤]¶V¨Ó¶V¤j¡A¦]¦¹¦p¦ó¥H¦UºØ§Þ³N¨Ó¹ï¦h´CÅé¸ê°T¶i¦æÀ˯Á»P¤ÀÃþ¡AÅܦ¨¬°¤@­Ó­«­nªº°ÝÃD¡C¥»­pµe¹w­p¬ã¨s¶}µo¥H¡u¥H¤º®e¬°¥Dªº»y­µ/­µ¼ÖÀ˯Á¨t²Î¡v¬°¥D¶bªº¬ÛÃö§Þ³N¡A­«ÂI±N©ñ¦b¦p¦ó¶}µo¾A·íªº¤ñ¹ï¾÷¨î¡A¨Ã¥B¦p¦ó°w¹ï¤j«¬¸ê°TÀ˯Á¨t²Î¨Ó¶i¦æ³Ì¨Î¤Æ¡A¥H¨Ï¤ñ¹ï®É¶¡¶Vµu¶V¦n¡C

  87. ¥H¤º®e¬°¥Dªº¦h´CÅéÀ˯Á¨t²Î ¢w ¥Ñ³nÅé¨ìµwÅ骺§Ö³tÂú§Î»Pµo®iÀô¹Ò

    • ­^¤å¦WºÙ: A Fast Prototyping Environment for Content-based Multimedia Information Retrieval
    • ­pµe½s¸¹:
    • ¥D«ù¤H: ±i´¼¬P
    • ¸É§U³æ¦ì: Üg«ä¬ì§Þ
    • ­pµe°õ¦æ´Á¶¡: 2000/12/1 to 2001/12/31
    • ÃöÁäµü: ¼Æ¦ì°T¸¹³B²z¡B¹Ï§Î¿ëÃÑ¡B¦h´CÅéÀ˯Á¡B¼v¹³³B²z¡B§Ö³tÂú§Î
    • ºK­n²¤¶:
      ÀHµÛ¹q¸£»Pºô¸ô¸ê°TªºÃz¬µ©Êµo®i¡A¦h´CÅé¸ê°TÀ˯Á§Þ³Nªº»Ý¨D»P¬D¾Ô¤]»P¤é­Ñ¼W¡C ¦h´CÅé¸ê°TÀ˯Á¬ã¨s¥²¶·µ²¦X¤£¦P»â°ìªºª¾ÃÑ»P§Þ³N¡A¥]¬A¼Æ¦ì°T¸¹³B²z¡B¼v¹³³B²z¡B­µ°T³B²z¡B¦ÛµM»y¨¥³B²z¡A¸ê®Æ®w§Þ³N¡Aºô¸ô§Þ³N¡AData Mining¡AAgentµ¥¡C¦Ó´¼¼z«¬¤H¾÷¤¶­±ªº»Ý¨D»PÀ³¥Î¼h¦¸¤]¶V¨Ó¶V¼sªx¡Cªñ¦~¨Ó¡A§ó¥Ñ©ósensor technologyªº¶i¨B¡A±a°Ê¤F¹Ï§Î¿ëÃÑ (Pattern Recognition) ªº¹ê¥Î¤Æ¡C¦]¦¹¡A¥H¿ëÃѧ޳N¬°°ò¦ªº´¼¼z«¬¦h´CÅéÀ˯Á©Î¿ëÃѨt²Î¤]ÀH³B¥i¨£¡A¨Ò¦p»y­µ¿ëÃÑ¡B»yªÌ¿ëÃÑ¡B¤HÁy¿ëÃÑ¡B¤â¼g¦r¿ëÃÑ¡BºqÁn¿ëÃÑ¡B¦W¤ù¿ëÃѵ¥¡C

      ¥»­pµe¹w­p¬ã¨s¶}µo¥H¡u¥H¤º®e¬°¥Dªº¦h´CÅéÀ˯Á¨t²Î¡v¬°¥D¶bªº¬ÛÃö§Þ³N¡A­«ÂI±N©ñ¦b¦p¦ó¥H²{¦³ªº MATLAB/Simulink/Toolboxes ¨Ó¶i¦æ¥Ñ³nÅé¨ìµwÅ骺§Ö³tÂú§Îµo®i¡]Fast Prototyping¡^¡A¨Ã¶}µo¥X¬ÛÃöªºµo®iÀô¹Ò»P¤H¾÷¤¶­±¡C§Ú­Ì±N¥H¦UºØ¦h´CÅéÀ˯Á©Î¿ëÃѨt²Î¨Ó¶i¦æ¶}µo»P´ú¸Õ¡A¨Ò¦p»yªÌ¿ëÃÑ¡]Speaker Recognition¡^¡B»y­µ¿ëÃÑ¡]Speech Recognition¡^¡B¤HÁy¿ëÃÑ¡]Face Recognition System¡^¡B¦±½Õ¿ëÃÑ¡]Melody Recognition¡^µ¥¡C

      ¥Ø«e¦h´CÅéÀ˯Á³£¬O¥H³nÅ鬰¥D¡A¦ý­Y­n±N¤§À³¥Î½d³òÂX¤j¡A«h»Ý¦Ò¼{±N¨äµwÅé¤Æ¡A¥H«K¯à¨Ï¥Î©ó¤é±`¥Í¬¡¡A¦p¤â¾÷¡B¢Þ¢Ò¢Ï¡A¥H¤Î¦UºØ®a¥Î¹q¾¹¥Î«~¡C±q³nÅé¨ìµwÅ骺¹Lµ{¡A°£¤F­n±Nºtºâªk§ï¬°¾A¦XµwÅé¹Bºâªº¤èªk¥~¡AÁÙ­n¦Ò¼{¦bµwÅé¤è­±ªº¦UºØ¹ê»Ú­­¨î¡A¨Ò¦p°O¾ÐÅ骺­­¨î¡B¥Ñ¯BÂI¼ÆÂà¦Ü©wÂI¼Æ©Ò³y¦¨ªº­pºâ»~®t¡B­pºâ³t«×ªº­°§C¡B¿é¥X/¿é¤Jªº¤¶­±­­¨îµ¥¡A³o¨Ç³£¬O§Ú­Ì­n¦Ò¼{ªº¦]¯À¡C

      ¦b¶Ç²Îªº§@ªk¤¤¡A¥Ñ³nÅé¦ÜµwÅ骺Âà´«¹Lµ{§¹¥þ­n¾a¤H¤u¨Ó§¹¦¨¡A¨ä¤¤¥]§t¤F³\¦h¡u¹Á¸Õ¿ù»~¡vªº¹Lµ{¡A¥»­pµe±N¦¹¹Lµ{ºÉ¶q¦Û°Ê¤Æ¡A§Q¥Î MATLAB/Simulink/Toolbox ªº¦UºØ²{¦¨¥\¯à¡A¹Á¸Õ±N¦b Simulink Àô¹Ò¤U©Òµo®iªº¿ëÃѨt²Î¡]¦p»yªÌ¿ëÃÑ¡B»y­µ©R¥O¿ëÃÑ¡B¦±½Õ¿ëÃÑ¡^Âà´«¨ì TI ªº 67xx ¨t¦CªºµwÅ饭¥x¡C¦¹¥~¡A§Ú­Ì¤]±Nµo®i¬ÛÃöªºµo®iÀô¹Ò»P¤¶­±¡A¹Á¸Õ±N Simulink ªº¦UºØ¿ëÃѨt²ÎÂà´«¦Ü§ó§C¶¥ªºª©¤l¡A¦p 8051 µ¥¡C

  88. ­µ¼ÖÀ˯Á§Þ³Nªº¥[³t»P§ï¨}

    • ­^¤å¦WºÙ: Efficient and Effective Techniques in Music Information Retrieval
    • ­pµe½s¸¹:
    • ¥D«ù¤H: ±i´¼¬P
    • ¸É§U³æ¦ì: ²M½«¬ì§Þ
    • ­pµe°õ¦æ´Á¶¡: 2000/9/1 to 2001/8/31
    • ÃöÁäµü: Music Information Retrieval, Multimedia Information Retrieval, Nearest Neighbor Search
    • ºK­n²¤¶:
      ÀHµÛ¹q¸£»Pºô¸ô¸ê°TªºÃz¬µ¡A¸ê°TÀ˯Á§Þ³Nªº»Ý¨D»P¬D¾Ô¤]»P¤é­Ñ¼W¡C¸ê°TÀ˯Á¬ã¨s¥²¶·µ²¦X¤£¦P»â°ìªºª¾ÃÑ»P§Þ³N¡A¥]¬A¦ÛµM»y¨¥³B²z¡A¸ê®Æ®w§Þ³N¡Aºô¸ô§Þ³N¡AData Mining¡AAgentµ¥¡C¦Ó´¼¼z«¬¤H¾÷¤¶­±ªº»Ý¨D»PÀ³¥Î¼h¦¸¤]¶V¨Ó¶V¼sªx¡Cªñ¦~¨Ó¡A§ó¥Ñ©ósensor technologyªº¶i¨B¡A±a°Ê¤F¹Ï§Î¿ëÃÑ (Pattern Recognition) ªº¹ê¥Î¤Æ¡C¦]¦¹¡A¥H¿ëÃѧ޳N¬°°ò¦ªº´¼¼z«¬¤H¾÷¤¶­±¤]ÀH³B¥i¨£¡A¨Ò¦p»y­µ¿ëÃÑ¡B»y­µ¦X¦¨¡B¤â¼g¦r¿ëÃÑ¡BºqÁn¿ëÃÑ¡B¦W¤ù¿ëÃѵ¥¡C

      ¥»­pµe¹w­p¬ã¨s¶}µo¥H¡u¥Hºq¿ïºq¡v¨t²Î¬°¥D¶bªº¬ÛÃöÀ˯Á»P¥[³t§Þ³N¡A¨Ã¾ã¦X»yªÌ¡B»y­µ¿ëÃѤ§§Þ³N¡A¨Ï¶}µo¥X¨Óªº´¼¼z«¬ÂIºq¨t²Î¡A¤£¦ý¯à±µ¨ü¨Ï¥ÎªÌªººqÁn¡B»y­µ¿é¤J¡A¨Ã¥ç¯à¦b§nÂøªºÀô¹Ò¤¤¡A¨³³t¥¿½T¦a§PÂ_ºqÁn¡B»y­µ¡B¯ºÁnµ¥¡A¨Ï¿ëÃѪºµ²ªG§ó¬°·Ç½T¡C¦¹¥~¡A§Ú­ÌÁÙ­n¬ã¨s¦UºØ¥[³t¤èªk¡A¨Ï¿ëÃѩҪ᪺®É¶¡¶V¤Ö¶V¦n¡A¨Ã¹Á¸Õ±N³o¨Ç¤èªk¹B¥Î©ó¤£¦P¥­¥x¤W¡]¦p Linux¡BWindows CE¡BPalm¡B®a¹q²£«~µ¥¡^¡A¥H¨Ï©Ò¶}µo¥X¨Óªº´¼¼z«¬ÂIºq¨t²Î¯àÀ³¥Î©ó¤é±`¥Í¬¡¤§¤¤¡C

      ¦b¥ý«eªº¬ÛÃö­pµe¤¤¡A§Ú­Ì¤w¸g¶}µo¥X¤@®M¡u¥Hºq¿ïºq¡vªº¨t²Î¡A¦b¦¹¨t²Î¤¤¡A¨Ï¥ÎªÌ¥i¥H¥Î¤K¬íªººqÁn¿é¤J¡]¤]¥i¥H¥i¥Î°Ûªº¡B­óªº¡A©Î¥Î¼Ö¾¹ºt«µ¡^¡AµM«á´N¯à¦b5-10¬í¤º¡A±q500­ººq¦±¤¤·j´M¥X³Ì¬Ûªñªº«e20¦W¡A¨Ã¨Ì¬Û¦ü«×¤j¤p±Æ¦C¡C¥»­pµe±N¥[³t­ì¥ýªºÂIºq¨t²Îªº¤ñ¹ïµ{§Ç¡A¨Ã¾ã¦X»yªÌµf¬q¿ëÃѤ§§Þ³N¡A¨Ï¶}µo¥X¨Óªº´¼¼z«¬ÂIºq¨t²Î¡A¤£¦ý¯à±µ¨ü¨Ï¥ÎªÌªººqÁn¡B»y­µ¿é¤J¡A¨Ã¥ç¯à¨³³t¦b§nÂøªºÀô¹Ò¤¤¡A¥¿½T§PÂ_ºqÁn¡B»y­µ¡B¯ºÁnµ¥¡A¥Î¥H±µ¨ü¨Ï¥ÎªÌ°T¸¹¤§¿é¤J¡C

  89. ²{¦³¼v¹³/»y­µ³B²z¬ã¨s¨å½d©óÂåÀø¦Û°Ê¤Æ¤§¯Ê¥¢»P§ï¶i - ¥H®ÖºÏ¦@®¶ÂåÀø»²§U¶EÂ_¨t²Î¤§¼v¹³³B²z§Y®É¤Æ¤Î¿é¤J»y­µ¤Æ¬°®×¨Ò

    • ­^¤å¦WºÙ: On-line Image Processing and Voice Activation in Magnetic Resonance Computer-Aided Diagnosis: A Strategy for Overcoming the Limitations in State of the Art Signal Processing Techniques
    • ­pµe½s¸¹:
    • ¥D«ù¤H: ³Å®a±Ò¡B´^®¶¿³¡B±i´¼¬P
    • ¸É§U³æ¦ì: °ê¬ì·|
    • ­pµe°õ¦æ´Á¶¡: 2000/8/1 to 2001/7/31
    • ÃöÁäµü: ÂåÀø¦Û°Ê¤Æ¡A¹q¸£»²§U¶q´ú¤Î¶EÂ_¡A®ÖºÏ¦@®¶¼v¹³¡A°ÊºA³W¹º¡A¦Û¾AÁ{¬É­Èªk¡B¼Ò½kC¥­§¡ªk¡A¤¤¶¡­È¤ÁÂ_ªk¡A»y­µ¿ëÃÑ¡A½u©Ê¹w´ú½s½X¡A­ËÀWÃСA°ÊºA®É¶¡®Õ·Ç¡AÁôÂæ¡°¨¥i¤Ò¼Ò«¬¡A³ÌÀu²Õ¦X·j´Mªk
    • ºK­n²¤¶:

      ¦ÛÂåÀø³]³Æ¼Æ¦ì¤Æ«á¡AÂåÀø¦Û°Ê¤Æ¬°¤@­«­n¤§¬ã¨s½ÒÃD¡A¦Ó¹q¸£»²§U¶EÂ_¬°ÂåÀø¦Û°Ê¤Æ¤§¤@¥D­n¤u¨ã¡AÁöµM¤w²Ö¿n¤j¶q¬ã¨s¦¨ªG¤Î¸gÅç¡A¬Æ¦Ü¤w¦³°Ó«~¤Æ¤§²£«~¡CµM¦Ó¤£½×¬O¼v¹³³B²z©Î¬O»y­µ¿ëÃÑ¡A¤@¯ë¦Ó¨¥¡A¦]¼v¹³³B²z¤§ºtºâ®É¶¡¹Lªø¤Î»y­µ¿ëÃѤ§®Ä«×¹L§C¡C¾É­P²{³õ¤§±µ¨ü«×¥¼¦p¹w´Á¡A¦P®É¥ç­­¨îÂåÀø¦Û°Ê¤ÆÂåÀø¬ã¨s¦¨ªG¹ï©óªÀ·|¤§°^Äm¡C

      ®ÖºÏ¦@®¶­º¦¸´£¨Ñ«D«I¤J©Ê·Ç½T´ú©w·s¦åºÞ¨t²Î¥¿±`¥\¯àªº¤èªk¡A¦]¨ä¬°«D«I¤J©Ê¶q´ú©ö³Q¨ü´úªÌ±µ¨ü¥B¹Ï¹³²M´·¡A¦]¦¹¬°¤@¬Æ¨ã¦³µo®i¼ç¤Oªº¶EÂ_²§±`¯fÅܤΫإ߹w¨¾©ÊÂå¾Ç¤§¤u¨ã¡C®ÖºÏ¦@®¶¹ï©óÀRºA¾¹©x¶EÂ_Áö¤w¨ú±o¼sªxªºÀ³¥Î¡A¦ý©|¥¼´¶¤Î©ó°ÊºA¾¹©x¡]¦p¤ß¦åºÞ¨t²Î¡^ªº¶EÂ_¡C°ÊºA¾¹©x¨ú¼Ë¸ê°T¶q»Ý»·¤j©óÀRºA¾¹©x¡A¦Ó²{¦³¤ß¤º(¥~)½¤Ãä¬É¤§ºtºâªkºtºâ®É¶¡¬Æªø¡AµLªkº¡¨¬§Y®É³B²z°ÊºA¾¹©x¤§¸ê°T¶q¡C

      ´N¹q¸£»²§U¶EÂ_¨t²Î¤§¨Ï¥ÎªÌ¦Ó¨¥¡A¦³®Ä»y­µ¨úÀɬ°¤@¬Æ¨üÅwªï¤§¥\¯à¤´¬Æ¬°»Ý­n¡A°ß¤@¤§­n¨D¬°°ª¿ëÃѲv¤Î°ªÃ­©w©Ê¡C²{¦³°Ó¥Î³nÅé¹ï©ó»y­µ¨úÀɤ§ÁZ®Ä¡A´ú¸Õµ²ªG¥¼¹F²{³õ¨Ï¥Î¤§­n¨D¡C­ì¦]¬°°Ó¥Î²£«~µo®i¤§¹ï¶H¬°«D¯S©w»yªÌ¤j¶q¦r·J¡A¦]¦Ó­°§C¿ëÃѲv¤Îí©w©Ê¡C´N¹q¸£»²§U¶EÂ_¨t²Î¦Ó¨¥¡Aµo®i¤@°ª¿ëÃѲv¤§¤p²³¤Ö¶q¦r·J»y­µ¿ëÃѺtºâªk¬°¤@¨Ï¹q¸£»²§U¶EÂ_¨t²Î§ó¨ã¤H©Ê¤Æ¤§¥D­n¥\¯à¡C

      ¦]ÂåÀø¦Û°Ê¤Æ¤§ºtºâªk¬°¤uµ{¬ã¨sªÌ©Ò¬ãµo¡A¶Ç²Î¤W±`°¾¦V¹êÅç«Ç¤uµ{¤H­û¤§Æ[ÂI¡A¥¼¥²¯à²Å¦X²{³õ¤§»Ý¨D¦Ó¾É­P§C±µ¨ü«×¡C¥»­pµe¤§¥Øªº¬°¥H²{³õ¨Ï¥ÎªÌ¤§»Ý¨Dµo®i¤@µo®i¤@§Y®É¤ß«Ç¤º¥~½¤Ãä¬ÉÀË´úºtºâªk¤Î»y­µ¨úÀɺtºâªk¡C¦p¥»­pµeÃÒ¹ê½T¹ê¥i¦æ¨Ã¬°²{³õ±µ¨ü¡A¤£¦ý¥i¹ï¶Ç²Î¤§¬ã¨s¨å½d«Ø¥ß¤@·sªº«ä¦Ò¤è¦V¡A¬ã¨s¤§¦¨ªG¥ç¥iª½±µ¦³¯q©óªÀ·|¤½²³¡C

  90. ¾ã¦X»y­µ¿ëÃÑ»P¦X¦¨ªº¥Hºq¿ïºq¨t²Î

    • ­^¤å¦WºÙ: On the Integration of Speech Recognition/Synthesis into a Content-based Music Retrieval System
    • ­pµe½s¸¹:
    • ¥D«ù¤H: ±i´¼¬P¡B¶À²ÐµØ
    • ¸É§U³æ¦ì: ²M½«¬ì§Þ
    • ­pµe°õ¦æ´Á¶¡: 2000/8/1 to 2001/7/31
    • ÃöÁäµü: »y­µ¿ëÃÑ¡B»y­µ¦X¦¨¡BºqÁn¿ëÃÑ¡B¼Æ¦ì°T¸¹³B²z¡B¹Ï§Î¿ëÃÑ
    • ºK­n²¤¶:

      ¥Ñ©ó¹q¸£»Pºô¸ô¬ì§Þªºµo¹F¡A´¼¼z«¬¤H¾÷¤¶­±ªº»Ý¨D»PÀ³¥Î¼h¦¸¤]¶V¨Ó¶V¼sªx¡Cªñ¦~¨Ó¡A§ó¥Ñ©ósensor technologyªº¶i¨B¡A±a°Ê¤F¹Ï§Î¿ëÃÑ (Pattern Recognition) ªº¹ê¥Î¤Æ¡A¦]¦¹¡A¥H¿ëÃѧ޳N¬°°ò¦ªº´¼¼z«¬¤H¾÷¤¶­±¤]ÀH³B¥i¨£¡A¨Ò¦p»y­µ¿ëÃÑ¡B»y­µ¦X¦¨¡B¤â¼g¦r¿ëÃÑ¡BºqÁn¿ëÃÑ¡B¦W¤ù¿ëÃѵ¥¡C

      ¥»­pµe¹w­p¬ã¨s¶}µo¥Hºq¿ïºq¨t²Î¡A¨Ã¾ã¦X»y­µ¿ëÃÑ»P»y­µ¦X¦¨¤§§Þ³N¡A¨Ï¶}µo¥X¨Óªº´¼¼z«¬ÂIºq¨t²Î¡A¤£¦ý¯à±µ¨ü¨Ï¥ÎªÌªººqÁn¿é¤J¡A¨Ã¥ç¯à±µ¨ü»y­µ°T¸¹ªº¿é¤J¡A¦Ó¥B¨t²Î¥»¨­¨ã¦³»y­µ¦X¦¨¤§¥\¯à¡A¯à°÷¥H»y­µ¦^ÂЦ³½ì¤§°T®§¡A©Î¥H»P¦]¤Þ¾Éªì¾ÇªÌ¦p¦ó¨Ï¥Î³o®M¨t²Î¡C

  91. ²MµØ¤j¾Ç¡u«D¦P¨B¤Þ¾É¦¡»·¶Z±Ð¾Ç¨t²Î¡v±À°Ê­pµe

    • ­^¤å¦WºÙ: Asynchronous Distance Learning Based on a Web Guiding System
    • ­pµe½s¸¹:
    • ¥D«ù¤H: ¤ý¤p¤t¡B¶À¤@¹A¡B±i´¼¬P¡B·¨¨û­ë
    • ¸É§U³æ¦ì: ±Ð¨|³¡
    • ­pµe°õ¦æ´Á¶¡: 2000/1/1 to 2000/12/31
    • ÃöÁäµü: Web-based learning, asynchronous distance learning
    • ºK­n²¤¶:
      ªñ¦~¡Aºô»Úºô¸ô(Internet)¤£¶È¦¨¬°¥þ¥@¬É³Ì¨üÅwªïªº´CÅé¡A¨Ã¦¨¬°°ê»Ú¶¡¸ê°TÄvª§ªº§Q¾¹¡Cºô»Úºô¸ôªºÀ³¥Î¤wµM¦¨¬°¿Å¶q¦U°ê¸ê°TÄvª§¤Oªº­«­n«ü¼Ð¡C³z¹LVOD (Video on Demand) ©M HTMLªº¤Þ¾É¦¡ºô­¶±Ð¾Ç¤è¦¡¡A¨äµL»·¥±©¡ªº¥\¯à¡A¨Ï±oºô¸ô±Ð¾Ç³vº¥¦¨¬°±Ð¾Çªº°t³Æ»P¼é¬y¡A¥¦¤£¶È»²¦õ¶Ç²Î±Ð«Ç²{³õ±Ð¾Ç¡A¤]¯à¿W¥ß«Øºc¬°µêÀÀ±Ð«Ç¡A¨Ï¾Ç¥Í¦³¦h¤¸ªº¿ï¾Ü¾Ç²ß¾÷·|¡A±µÄ²¸ê°TªÀ·|ªº¾Ç²ßÀô¹Ò¡C ¨Ï¥ÎVODªº¦n³B¬O¤ñ¸û¨ã¯u¹ê©Ê¡A³z¹LVideo¡A§Ï©»¦b±Ð«Ç¤W½Ò¤@¼Ë¡A¦Ñ®v±Â½ÒªºÁn­µ¼v¹³µ¥¶Ç±Â¤è¦¡³£¥i¥ß§Y·P¨ü¨ì¡A¦¹¥~¡A¦Ñ®vªº±Ð§÷¤]¤£»Ý¨Æ¥ý¯S§O¦w±Æ©Î³B²z¡A¦Ñ®v¤£¶È¥i¥H¹³¥­®É¦b±Ð«Ç¤¤±Ð¾Ç¤@¼Ë¡A¨Ï¥Î§ë¼v¤ù¡A¤]¥i§Q¥Î¥ÕªO¸Ñ»¡(¦p¾É¼Æ¾Ç¤½¦¡µ¥)¡F¦ý¨ä¯ÊÂI¬O»Ý¨Ï¥Î¸û°ªªºÀW¼e¡A¦P®É¹ï©ó¤å¦rªº¸ÑªR«×µy®t¡C¦Ü©ó¨Ï¥ÎHTMLªº¦n³B´N¬O¨Ï¥ÎªºÀW¼e¸û§C¡A¤å¦r¤è­±ªº¸ÑªR¯à¤O¤]¸û¥O¤Hº¡·N¡A¤×¨ä¦b¾Ç²ß¤W¡A¤ñ¸û®e©ö¤ÏÂо\Ū¡A¤£¹³¥ÎVideo¦³³sÄò¼½©ñªº¾Ç²ßÀ£¤O¡C¤£¹L¡A¨Ï¥ÎHTML«o¸û©ö»P¦Ñ®v¦³²¨Â÷·P¡A¤£¨ã¯u¹ê©Ê¡A¦Ñ®v¤]¬°¤F°t¦XHTMLªº®i²{¤è¦¡¡A»ÝÃB¥~ªá®É¶¡»P¤H¤O»s§@±Ð§÷¡C³Ì·sªººô¸ô±Ð¾Ç¤è¦¡--¤Þ¾É¦¡ºô­¶±Ð¾Ç¡A¬Oºî¦XVOD©M HTMLªº§Þ³N¡A¤¬¸É©¼¦¹ªºÀu¯ÊÂI¡A±Â½Ò±Ð®v±N±Ð§÷©Î¿ý¼v±a¥Hºô­¶ªº§Î¦¡©ñ¦bºô¯¸¤W¡A¾Ç¥Í¥iÀH®É³z¹Lºô¸ô¾\Ū±Ð§÷¡A¨Ã°Ñ»P°Q½×¡A¨Ï±o±Ð®v»P¾Ç¥Í¦³Âù¦V·¾³qªº¾÷·|¡C¦p¦¹ªº¾Ç²ß¤è¦¡¡A°£¥i´£¨Ñ¸û¦hªº¦h¤¸¤Æ¶i­×ºÞ¹D¥~¡A¥ç¥i´£¤É±Ð¾Ç«~½è¡A¹F¨ì¾Ç³N¥æ¬y¡B¸ê·½¦@¨Éªº¥Øªº¡C¹w´Á±N¯à§ó¦³®Ä¦a´£¤É¶Ç²Î±Ð¾Çªº®Ä¯à¡A¨ÃÂX¤j±Ð¨|¶Ç¼½ªº½dÃ¥¤Î¯à¤O¡C¤Q¦~¾ð¤ì¡A¦Ê¦~¾ð¤H¡C±Ð¨|Àô¹Òªº§ï¶i¡A¼vÅT²`»·¡CÂǥѺô¸ô»·¶Z±Ð¾ÇÀô¹Òªº¦¨¥ß¡A§Ú°êNII¬Fµ¦¹ªÀy¥þ¥Á¤Wºô¡A±Ð¨|³¡­Ò¾É¡u²×¨­¾Ç²ß¡v¡A³£¥iÂǦ¹«Å´­»P¸¨¹ê¡C

      ¥»®Õ¦Û85¦~°_§Y¿n·¥°Ñ»P»·¶Z±Ð¾Ç­pµe¡Aºô¸ô±Ð¾Ç¬ÛÃö§Þ³N»P®v¸ê³£¦b°ö¾i¤¤¡A¦UÃþºô¸ô±Ð¾Ç¤è¦¡»P¦¨ªG³£³vº¥Åã²{¡A­pºâ¾÷¤¤¤ß¤£Â_¦a´£¨Ñ§ï¶i¦UÃþ§Þ³N¤ä´©¡A§Y±N¤Þ¶iªº·s§Þ³N¡y¤Þ¾É¦¡ºô¸ô±Ð¾Ç¨t²Î¡z¡A¬O°ê¤º¦U¤j¾Ç¤§­º³Ð¡AÂǥѥ¦ªºÀ°§U¥iÅý±Ð®v§ó§¹¾ã¥B²M´·ªº¶Ç¹F±Â½Òªº¤º®e¡A¾Ç¥Í­Ì¤]¦]¦Ó¦³§ó¦nªº¾Ç²ßÀô¹Ò¡C¦¹¥~¡A­pºâ¾÷¤¤¤ßÁÙ­pµe´£¨Ñ§ó¦hªººô¸ô±Ð¾Ç¤u¨ã¨Ó¨ó§U¦Ñ®v¹Á¸Õ¡CÅý¦Ñ®v±a»â¾Ç¥Í¤@°_ªï¦V¸ó¥@¬öªº¸ê°T·s¥@¥N¡A¦bºô¸ôªÀ·|¤¤«Ø³y³Ð·s¥B¨ã¬D¾ÔªºÀu¶Õ¡C¬ÛÃö¸ê®Æ»P¦¨ªG¥i¨£¥»®Õ­º­¶¡Ghttp://www.nthu.edu.tw/ ¤Î«D¦P¨B±Ð¾Ç­º­¶¡Ghttp://sailnet.eden.nthu.edu.tw/¡C

  92. ¥HÁn¯¾»P¤HÁy¬°¥Dªº¥Íª«»{µý¨t²Î

    • ­^¤å¦WºÙ: Biometric Identification System Based on Face and Voice
    • ­pµe½s¸¹:
    • ¥D«ù¤H: ±i´¼¬P
    • ¸É§U³æ¦ì: ¥îÂ׬ì§Þ
    • ­pµe°õ¦æ´Á¶¡: 1999/10/1 to 2000/7/31
    • ÃöÁäµü: ¥Íª«»{ÃÒ¡B¹Ï§Î¿ëÃÑ¡BÁn¯¾¤ñ¹ï¡B»yªÌ¿ëÃÑ¡B¤HÁy¿ëÃÑ¡B¼Æ¦ì°T¸¹³B²z¡B¼v¹³³B²z
    • ºK­n²¤¶:

      ¥Ñ©ó¹q¸£»P ºô¸ô¬ì§Þªºµo¹F¡A»{ÃÒ (Authentication) ªº»Ý¨D»PÀ³¥Î¼h¦¸¤]¶V¨Ó¶V¼sªx¡Cªñ¦~¨Ó¡A§ó¥Ñ©ósensor technologyªº¶i¨B¡A±a°Ê¤F¥Íª«»{ÃÒ (Biometric identification) ªº¹ê¥Î¤Æ¡A³o¬O¾Ç³N¬É¤Î¤u°Ó¬É¬Æ¬°­«µøªº¤@ºØ¦w¥þ»{ÃҤ覡¡A¨ä»{ÃҩҥΨ쪺¥Íª«¯S¼x¥]§t«ü¯¾¡B´x¯¾¡B»y­µ¡B¤HÁy¡B­i½¤¡B¤â«¬¡B¦Õ«¬¡B¼L®B¹B°Ê¼Ò¦¡µ¥¡A¦¹ºØ»{ÃÒªºÀuÂI¬°°°³y¤Î¯}¸Ñ§xÃø¡B¨Ï¥Î¤è«K¡]¤£¾á¤ß¿ò¥¢¡^¡B¾A¥Î©Ê¼sªx¡]¥i¨Ì¦w¥þ©Ê»Ý¨D¨Ó½Õ¾ã¡^¡C¦Ó¯ÊÂI«h¦b©ó§Þ³N¼h¦¸¸û°ª¡B»Ý­n¸û¦hªº³nµwÅé¤ä´©¡A¥B»Ý­n©Mµ½ªº¤H¾÷¤¶­±¡A¥H¥R¤Àµo´§¨ä¥\¯à¡C

      ¥»­pµe¹w­p¬ã¨s¶}µo¥HÁn¯¾»P¤HÁy¬°°ò¦ªº¥Íª«»{ÃÒ¨t²Î¡A§Æ±æ¯àÀ³¥Î²{¦³ªº³nµwÅé§Þ³N¡A¶}µo¥Xí©T¡B¥i¥Îªº³nÅé¨t²Î­ì«¬¡A¥H¬°¥Íª«»{ÃÒªº¹ê¥Î¤Æ¡B´¶¤Î¤Æ»P°Ó«~¤Æ¶i¦æ¥ý¾É¬ã¨s»P±´°É¡C

  93. ª½¬yÅÜÀWªÅ½Õ¾÷±±¨îµ¦²¤µo®i

    • ­^¤å¦WºÙ: Design and developments of the control strategy for a DC variable-frequency air conditioner
    • ­pµe½s¸¹:
    • ¥D«ù¤H: ±i´¼¬P
    • ¸É§U³æ¦ì: ¤u¬ã°|¯à¸ê©Ò
    • ­pµe°õ¦æ´Á¶¡: 1999/10/1 to 2000/10/31
    • ÃöÁäµü: ¨t²Î¿ëÃÑ¡B±±¨îµ¦²¤¡BPID±±¨î¡BFuzzy±±¨î¡B¹q¸£»²§U±±¨î¨t²Î³]­p
    • ºK­n²¤¶:

      ª½¬yÅÜÀWªÅ½Õ¾÷¬O¤@­Ó¨å«¬ªº«D½u©Ê®ÉÅܨt²Î¡A¨ä¯S©ÊÀHµÛÀ£ÁYÀW²vªº¤£¦P¦Ó¦³¤£¦PªºÅܤơA¦]¦¹¦b¨t²Î¿ëÃѤα±¨îµ¦²¤¤è­±³y¦¨¸û°ªªº§xÃø«×¡CµM¦Ó¡Aª½¬yÅÜÀWªÅ½Õ¾÷ªº±±¨î¤@ª½¬O¤@­Ó«Ü­«­nªº½ÒÃD¡A¦¹¥i¥Ñ¤@¯ë¤é¨t®a¹q²£«~¡]¯S§O¬O§N®ð¾÷¡^©Ò±j½Õªº¡u´¼¼z«¬±±¨î¡v©Î¡uNeuro-fuzzy control¡v¦Ó²¤¿s¤@¤G¡C

      ¥»­pµe±N¥Hº¥¶i¤è¦¡´`§Ç¹Á¸Õª½¬yÅÜÀWªÅ½Õ¾÷ªº¨t²Î¿ëÃÑ»P±±¨îµ¦²¤¡Cªì´Á±N¥H¶Ç²Îªº½u©Ê¨t²Î¿ëÃѤÎPID±±¨î¬°¥D¡A«Ý²Ö¿n°÷¦h¸gÅç«á¡A±N¶i¤@¨B¹Á¸Õ¥HGain schedulingªº¤è¦¡¨Ó¶i¦æ«D½u©Ê±±¨î¡A³Ì«á¦A¥HFuzzy±±¨î¨Ó¹F¨ì¾ã­Ó¨t²Îªº³Ì¨Î¤Æ¡C

  94. ¯«¸g¼Ò½k¨t²Îªº»~®t¹w¦ô»Pµ²ºc¿ëÃÑ

    • ­^¤å¦WºÙ: Error Estimation and Structure Identification of Neuro-fuzzy Systems
    • ­pµe½s¸¹: NSC 89-2213-E-007-067
    • ¥D«ù¤H: ±i´¼¬P
    • ¸É§U³æ¦ì: °ê¬ì·|
    • ­pµe°õ¦æ´Á¶¡: 1999/8/1 to 2000/7/31
    • ÃöÁäµü: ³n¦¡­pºâ¡]Soft Computing¡^¡B­pºâ´¼¯à¡]Computational Intelligence¡^¡B¨t²Î¿ëÃÑ¡]System Identification¡^¡B³Ì¤p¥­¤èªk¡]Least-squares Estimator¡^¡B»¼°j¦¡³Ì¤p¥­¤èªk¡]Recursive Least-squares Estimator¡^¡B¸ê®Æ¼Ò«¬¤Æ¡]Data Modeling¡^¡B»~®t¹w¦ô¡]Error Measure Estimation¡^¡Bµ²ºc¿ëÃÑ¡]Structure Identification¡^¡BÃþ¯«¸gºô¸ô¡]Artificial Neural Networks¡^¡B¼Ò½kÅÞ¿è¡]Fuzzy Logic¡^¡B¯«¸g¼Ò½k¨t²Î¡]Neuro-fuzzy Systems¡^¡B¿ò¶Çºtºâªk¡]Genetic Algorithms¡^¡B¼ÒÀÀ°h¤õªk¡]Simulated Annealing¡^¡B¹Ï«¬ÃѧO¡]Pattern Recognition¡^¡B¸ê®Æ±´°É¡]Data Mining¡^¡Bª¾Ãѵo±¸¡]Knowledge Discovery¡^
    • ºK­n²¤¶:
      ¦b¤W¦~«×ªº°ê¬ì·|­pµe¡]"³n¦¡­pºâªº»~®t¹w¦ô»Pµ²ºc¿ëÃÑ"¡^¤¤¡A§Ú­Ì§¹¦¨¤F¤U¦C¨Æ¶µ¡G

      1. ¥H»¼°j¦¡³Ì¤p¥­¤èªk¡]Recursive least-squares estimator¡A²ºÙRLSE) ¬°°ò¦¡Aµo®i¥X­pºâ leave-one-out error ªº incremental formula¡A¥i¥Î©ó neuro-fuzzy modeling ªºµ²ºc¿ëÃÑ¡]Structure identification¡^¤Îª[²§ÂIªº°»´ú¡]Outlier detection¡^
      2. ¥H leave-one-out error ¬°°ò¦¡A¦A¤Þ¥Î sequential forward input selection ©Î exhaustive input selection¡A¨Ó¶i¦æ«ÈÆ[¥B¨³³tªº¿é¤J¿ï¨ú¤Îµ²ºc¿ëÃÑ¡C

      ¦³Ãö©óleave-one-out errorªº incremental formula¥H¤Î¬ÛÃöªº¿é¤JÅܼƿï¨ú¤èªk¡A¤wµoªí¦b1998¦~5¤ëªºIEEE International Conference on Fuzzy Systems [Jang98fuzz]¡AÃþ¦üªº¤èªk¥ç¥i¥Î©óª[²§ÂIªº°»´ú¡]Outlier detection¡^¡A¤wµoªí¦b1998¦~10¤ëªºInternational Joint Conference on Information Science [Jang98jcis]¡C§¹¾ãªº±À¾É»P¬ÛÃöªºÀ³¥Î«h¤w¥æ¥I IEEE Transactions on Fuzzy Systems ¼f¬d¤¤[Jang98itfs]¡C¥»¦~«×¤§°ê¬ì·|­pµe¤D¦b©µÄò¤W¦~«×¤§­pµe¡A¨ä¥Øªº¦³¤T¡G

      1. ©µÄò¨ÃÀ³¥Î¤W¦~«×¤§¬ã¨s¤ß±o¡A¶i¤@¨B¨Ï¥Î¥­¤è®Úª©¡]Square-root version¡^ªº»¼°j¦¡³Ì¤p¥­¤èªk¨Ó¶i¦æ¯«¸g¼Ò½k¼Ò«¬¡]Neuro-fuzzy models¡^ªºleave-one-out error ªº­pºâ¡A¥HºÉ¶q­°§C¤j¶q¼Æ­È¹Bºâ¤Î¹q¸£¤º³¡¦³­­ºë·Ç«×©Ò±a¨Óªº±Ë¥h¡]Truncation¡^¤Î¶i¦ì¡]Round-off¡^»~®t¡C
      2. ¶i¤@¨B¹Á¸Õ¦UºØµ²ºc¿ëÃѪº¤èªk¡A¥]§tGMDH (Group Method of Data Handling) [] ªº¤èªk¡A¥H¤Î¨Ï¥Î©ó«Øºc¦¡¾Ç²ß¡]Constructive learning¡^¤Î¯}Ãa¦¡¾Ç²ß¡]Destructive learning¡^ªºIncremental formulas¡C¡]¦¹Ãþformulas ¥iµø¬°recursive least-squares estimator in the number of parameters¡A¦]¦¹¬Û·í¨³³t¡C­Y¦A¥[¤W«e­z square-root version of LSE¡A«h­pºâ¥i»¡¬O¤S§Ö¤S·Ç¡C¡^
      3. Ä~ÄòºûÅ@Working Group on Data Modelingªº­º­¶¡]¦¹¬°¦bIEEE Neural Networks Council Standards Committee¤§¤UªºSubcommittee¡A¥Ñ Texas A & M University ¸ê°T¨tªºProf. John Yen©e°U§Ú¥»¤H©ó1996¦~©³¶}©l«Ø¸m¡^¡A¨Ã¥[±j¨ä¤º®e»PªA°È¡C

      ³n¦¡­pºâ¡]Soft computing¡^¬Oªñ¦~¨ÓProfessor Lotfi Zadeh¡]¼Ò½kÅ޿褧¤÷¡AU.C. BerkeleyªººaÅA°h¥ð±Ð±Â¡^´£­Òªº·s¤è¦V¡A¨äºë¯«¤D¬Oµ²¦XÃþ¯«¸gºô¸ô¡]Artificial neural networks¡^¤Î¼Ò½kÅÞ¿è¡]Fuzzy logic¡^ªºÀuÂI¡A¨Ã»²¥H¤£¶·¾É¦¡ªº³Ì¨Î¤Æ¡]Derivative-free optimization¡^¤èªk¡A¨Ó°w¹ï¨ú¼Ë¸ê®Æ¡]Sample data¡^¤Î¬J¦³ªº±M®aª¾ÃÑ¡]Expert knowledge¡^¶i¦æ¼Ò«¬ªº¤ÀªR¤Î·L½Õ¡]Fine tuning¡^¡A¥H«Ø¥ß¤@­Ó¨ã¦³¾Ç²ß¯à¤Oªº´¼¼z«¬¨t²Î¡A¨Ã¥i¹ï©ó®ÉÅÜ¡]Time-varying¡^ªºÀô¹Ò¶i¦æ¦Û§Ú§Y®É½Õ¾A¡]On-line adaptation¡^¡A¥HÀò¨ú³Ì¨Îªº®Ä¯à¡C

      ¦b¼Ò«¬ªº¿ï¨ú¤W¡A³n¦¡­pºâ¬O°¾¦V©ó¨Ï¥ÎÃþ¯«¸gºô¸ô©Î¼Ò½k±Àºt¨t²Î³o¨âÃþ¼Ò«¬¡C¥Ñ©óÃþ¯«¸gºô¸ô¬O¨ã¦³¾Ç²ß©Î½Õ¾A¯à¤O¡]Learning or adaptation capability¡^ªº¶Â½c¼Ò«¬¡]Black-box model¡^¡A¦Ó¼Ò½k±Àºt¨t²Î¡]Fuzzy inference systems¡^«h¬O¯àªí¥Ü±M®aª¾ÃѪº¼Ò½k³W«h®w¨t²Î¡]Fuzzy rule-based systems¡^¡A¦]¦¹³n¦¡­pºâ¯S§O±j½Õ³o¨âªÌªºµ²¦X¡A§Î¦¨­Ý¨ã¨âªÌ¤§ªøªº¯«¸g¼Ò½k±Àºt¨t²Î¡]Neuro-fuzzy inference systems¡^¡A¨äÀ³¥Î½d³ò¬Û·í¼sªx¡AÁ|¤Z¹ï©ó¸ê®Æ©Î±M®aª¾ÃѪº¼Ò«¬¤Æ¡]Modeling¡^¡A§¡¥i¥Î±o¤W¡CµM¦Ó¦b¹ê»ÚªºÀ³¥Î¤W¡A¤´¦³³\¦h°ÝÃD«E«Ý§JªA¡A¨Ò¦p¿é¤JÅܼƪº¿ï¨ú¡]Input selection¡^©MÅܧΡ]Transformation¡^¡B¿é¤JªÅ¶¡ªº¤À³Î¡]Input space partitioning¡^¡B¼Ò½k³W«h¼Æ¥Øªº¿ï©w¡B¯}Ãa¦¡¤Î«Øºc¦¡ªº¾Ç²ß¡]Destructive and constructive learning¡^µ¥µ¥¡A³o¨Ç°ÝÃD³£¬O¦b¶i¦æµ²ºc¿ëÃÑ¡]Structure identification¡^®É©Ò»Ý¸Ñ¨Mªº°ÝÃD¡A¤]³£¬O¥»­pµeªº¬ã¨s­«ÂI¡C¡]³o¨Ç°ÝÃD¦b«e¤@¦~«×ªº­pµe¤w¦³ªì¨B¦¨ªG¡A¦ý¤´¦³³\¦h potential approaches ©|¥¼±´°Q¡A¥Î©ó­pºâ leave-one-out error ªº¥­¤è®Úª©ªº»¼°j¦¡³Ì¤p¥­¤èªk¤]©|¥¼±À¾É§¹¥þ¡C¡^

      ¯«¸g¼Ò½k¨t²Îªº°ò¥»¾Ç²ß¤èªk¬°°f¶Ç¾Éªk¡]Back-propagation¡^¡A§Y¬°Â²³æªº±è«×¤U­°ªk¡]Gradient descent¡^©Î¬O³Ì³t¤U­°ªk¡]Steepest descent¡^¡A§ó½ÆÂø¥ý¶iªº¤èªk«h¬O¦b²Î­p©Î«D½u©Ê°jÂk¡]Nonlinear regression¡^¤¤±`¥Î¨ìªºGauss-Newton Method©Î¬OLevenberg-Marquardt Method¡C¦ý¬O³o¨Ç¤èªk³£¶·­n¥Î¨ì±è«×¦V¶q¡]Gradient vectors¡^¦Ó±è«×¦V¶q¦b½ÆÂø¨t²Î¤¤¨Ã¤£®e©ö­pºâ¡A¦]¦¹¹ï©ó¸û½ÆÂøªº¤j«¬¨t²Î¡A³n¦¡­pºâ°¾¦V©ó¨Ï¥Î¤£¶·¾É¦¡ªº³Ì¨Î¤Æ¤èªk¡]Derivative-free optimization¡^[Jang97book]¡A¨Ò¦p¿ò¶Çºtºâªk¡]Genetic algorithms¡^¡B¼ÒÀÀ°h¤õªk¡]Simulated annealing¡^¡B¤U©Y¦¡Simplex·j´M¡]Downhill simplex search¡^¡BÂø¶Ã·j´M¡]Random search¡^¡B¥¦¥¬·j´M¡]Tabu search¡^µ¥¡C¦b§Ú­Ìªº¬ã¨s¤¤¡A¬O©T©w«D½u©Ê°Ñ¼Æ¡A¦ÓºÉ¶q±Ä¥Î³Ì¤p¥­¤èªk¨Ó¿ëÃѽu©Ê°Ñ¼Æ¡A¨ä¦n³B¬°­pºâ¨³³t¡A¥B mathematically tractable¡C­Y­pºâ¸ê·½Â×´I¡A«h¥ç¥i¦Ò¼{¥H¤W­z«D½u©Êªº¤èªk©Î¤£»Ý¾É¼Æªº¤èªk¨Ó¶i¦æ«D½u©Ê°Ñ¼Æªº·L½Õ¡C

      ¬°´£°ª¥xÆW¦b¦¹¤è­±¬ã¨sªºª¾¦W«×¡A¨Ã«P¶i§Þ³N¤Î¸ê®Æªº¥æ¬y¡A¥»­pµe±NÄ~ÄòºûÅ@Working Group on Data Modelingªº­º­¶¡A¦¹¬°¦bIEEE Neural Networks Council Standards Committee¤§¤UªºSubcommittee¡A¥ÑProf. John Yen©e°U¥»¤H©ó1996¦~©³¶}©l«Ø¸m¡AURL¦a§}¬°¡Ghttp://www.cs.nthu.edu.tw/~jang/benchmark¡C¥»­pµe¬ã¨s¶}µoªº¦¨ªG¡A¤]§Æ±æ¯à°÷¥H CGI (Common Gate Interface) ªº¤è¦¡¡A¸m©ó¦¹ºô­¶¤W¡A¥H¨Ñ¦³¿³½ì¤§¬ã¨s¦P¦n¤W¶Ç Training data ¤Î Test data¡A¥Ñ¦øªA¾¹°õ¦æ»~®t¹w¦ô¤Îµ²ºc¿ëÃÑ«á¡A¦b¥ß¨è±Nµ²ªG¶Ç¦^ Client ºÝ¡A¥H¨Ñ¬ã¨s¦P¦n¶i¦æ¤ñ¸û»P´ú¸Õ¡C

  95. ·s»D¤ÀÃþ»P¤å¥óºK­n§Þ³N¶}µo

    • ­^¤å¦WºÙ: Classification and Summarization for On-line News
    • ­pµe½s¸¹:
    • ¥D«ù¤H: ±i´¼¬P
    • ¸É§U³æ¦ì: ¤Ó¤@«H³q
    • ­pµe°õ¦æ´Á¶¡: 1999/8/1 to 2000/7/31
    • ÃöÁäµü: ¤Ó¤@·s»D¡B¤å¥ó¤ÀÃþ¡B·s»D¤ÀÃþ¡Bºô­¶¦Û°Ê¥Í¦¨¡B¤å¥óºK­n¡B¤¤¤å¦r µü»y®Æ®w¡B»y·N¤ÀªR
    • ºK­n²¤¶:

      ¤Ó¤@·s»D¥Ø«e©Ò´£¨Ñªººô¸ô·s»DªA°È¬OÂ^¨ú¦h­Óºô¯¸ªº·s»D¸ê®Æ¡A¨Ã±N¨ä¤Àªù §OÃþ¡A°£¤F¤è«KŪªÌ®Ú¾ÚÃþ§O¨Ó¾\Ū¡A¦P®É¤]¥i¥H¤ñ¸û¦h®a·s»Dºô¯¸¹ï¦P¤@¨Æ¥óªº³ø¾É¡C°w¹ï·s»D¤ÀÃþ³o¤@¶µ¤u§@¡A¦pªG¯à°÷¼W¥[¹q¸£¤ÀÃþªººë½T«×¨Ï¨ä¹F¨ì§Ú­Ì¹w´Áªº¤ÀÃþµ²ªG¡A¥B¹Lµ{¤¤¤£»Ý­n¤H­û¤¶¤J¨ó§U¡A±N¥i¤j¶q¦a¸`¬Ù¦¹¶µ¤u§@©Òªáªº®É¶¡¡C

      ¦Óºô»Úºô¸ôªºµo®i¤@¤é¤d¨½¡Aºô¯¸¤Wªº¤å³¹¸ê°T¤]¦h¦pÁc¬P¡C·í¨Ï¥ÎªÌ­±¹ïºô¸ô¤W»P¤é­Ñ¼Wªº¤å³¹®É¡A±`¦³¤£ª¾±q¦ó§ä°_¤§¾Ñ¡CÁöµM¥Ø«e¤w¸g¦³³\¦h·j´M¤ÞÀº¨ÓÅý¨Ï¥ÎªÌ®e©ö¦bºô»Úºô¸ô¤W·j´M¾A¥Îªº¤å¥ó¡A¦ý·j´M¤ÞÀº©Ò´£¨ÑªºÂ²¤¶©Î¬O¹Lµu¡A¬ÆªÌ¦³»~¾É¤§ºÃ¡C¦]¦¹»Ý­n¬°ºô¸ô¤W¤å¥ó»s§@¥¿½TªººK­n¡A¦³®Ä¦a¬°¨Ï¥ÎªÌ´£¨Ñ©Ò»Ýªº¸ê°T¡C

      ¥»­pµe¹w­p¬ã¨s¶}µo¤@¨t¦Cªººô¸ô¸ê·½±´°É¡B¤å¦r¤ÀªR¡Bºô­¶µ{¦¡³]­p¡A¥H¤Î¦hºØ¤ÀÃþ¤ÎºK­n§Þ¥©¡A¥H«K©ó´£¨Ñ¤å¥óºK­n©M·s»D¤ÀÃþªº§¹µ½¸Ñ¨M¤è®×¡C

  96. ¸õÀWºô¥x¤ÀªR

    • ­^¤å¦WºÙ: The analysis of radio communication networks
    • ­pµe½s¸¹:
    • ¥D«ù¤H: ±i´¼¬P
    • ¸É§U³æ¦ì: ¤¤¬ì°|
    • ­pµe°õ¦æ´Á¶¡: 1999/7/1 to 2001/6/30
    • ÃöÁäµü: »EÃþ¤ÀªR¡B¹Ï§Î¿ëÃÑ¡BµL½u¹qºô¥x°T¸¹¤ÀªR
    • ºK­n²¤¶:

      µL½u¹qºô¥x¤ÀªR¾¹¡]Radio Network Analyzer¡^ªº¥ô°È¬O±qºI¦¬¨ìªºµL½u¹q°T¸¹¤¤¡A³v¤@¿ëÃÑ¥X¬Û¹ï¬M¤§µL½u¹q³q«H¾÷¤§µo¸Ü¡BµL½u¹q³q«H¾÷¡A¥H¤ÎµL½u¹qºô¡C¦b¦¹´£¤Îªº¦U¶µ¦Wµü¦s¦b¤@­Ó¶¥¼h¦¡ªºÃö«Y¡G

      1. µL½u¹qºô¬O¥Ñ¼Æ­ÓµL½u¹q³q«H¾÷©Ò²Õ¦¨
      2. µL½u¹q³q«H¾÷¥i¶i¦æ¤£­­©w¦¸¼Æ¤§µo¸Ü
      3. ¨C¤@­Óµo¸Ü¬O¥Ñ¤@³s¦êªºhop©Ò²Õ¦¨

      ºI¦¬¾¹°O¿ý¤U¨ÓªºµL½u¹q°T¸¹§Y¬O¤@³s¦êªºhop¡A¦ÓµL½uºô¥x¤ÀªR¾¹ªº¥\¯à§Y¬O±NhopÂkÃþ¦Ü¥i¯à¤§µo¸Ü¡A¨Ã¶i¦Ó¤ÀªR­pºâ¥X¦¹µo¸Ü¥i¯à·½¦Û¬Y¤@µL½u¹q³q«H¾÷¡A¥H¤Î¦¹µL½u¹q³q«H¾÷¥i¯àÁõÄݪºµL½u¹qºô¡C¦]¦¹µL½u¹qºô¥x¤ÀªR¥i¤À¬°¤T¨BÆJ¡G

      1. µo¸Ü¤ÀªR¡G±Nhop ÂkÃþ¨ì¬Y¤@µo¸Ü
      2. µL½u¹q³q«H¾÷¤ÀªR¡G±Nµo¸ÜÂkÃþ¨ì¬Y¤@µL½u¹q³q«H¾÷
      3. µL½u¹qºô¤ÀªR¡G±NµL½u¹q³q«H¾÷Âk¨ì¬Y¤@µL½u¹qºô

      ¦b³o¤T­Ó¨BÆJ¤¤¡A¦³¤@­Ó¦@¦Pªº¤ÀªR¤èªk¡A´N¬O¥²¶·±N¸ê®Æ¥[¥H»E¶°¤Î¤ÀÃþ¡A¦]¦¹¥»­pµeªº­º­n¥Ø¼Ð§Y¬O¹Á¸Õ¥H¦UºØ¸s»E¤ÀªR(Clustering Analysis) ¤Î¹Ï§Î¿ëÃÑ(Pattern Recognition)ªº¤èªk¡A¨Ó¶i¦æµL½u¹qºô¥x¤ÀªR¡C

      ¦b«e¤@¦~ªº­pµe¡]µL½u¹q³q«H¤§ºô¥x¤ÀªR¡^¤¤¡A§Ú­Ì¤w¸g§¹¦¨¤F¤@­Ó°ò¥»³nÅé¡]µL½u¹qºô¥x¤ÀªR¤u¨ã½c, Radio Network Analysis Toolbox)¡A¥i¹ïµL½u¹q³q«Hºô¶i¦æ°ò¥»ªººô¥x¿ëÃÑ»P©w¦ì¡C¦¹¤u¨ã½c¥i¦b¨âºô¤T¥xªº±¡ªp¤U¡A°w¹ï¨â­ÓºI¦¬¾¹ªº¸ê®Æ¡A¶i¦æ¤U¦C¦UºØºô¥x¤ÀªR¡A³Ì«á±o¨ì¦U­Ó³q°T¾÷ªº¦ì¸m¸ê°T¡C¦b¥»¦~«×ªº­pµe¤¤¡A§Æ±æ¯à§ï¶i¦¹¤u¨ã½c¡A¨Ï¨ä¥\¯à§ó¬°±j¤j¡A¥D­n¥Ø¼Ð¦³¤U¦C¦UÂI¡G

      1. ¥i¾A¥Î©ó¦hºô¦h¥x¡]²Ä¤@¦~¹w´Á¥Ø¼Ð¬°¤­ºô¤­¥x¡A²Ä¤G¦~¹w´Á¥Ø¼Ð¬°¤Kºô¤K¥x¡^
      2. ¥i¦Û°Ê¤Æ¦a¿ëÃѳq°T¾÷ÅÜ´«¦ì¸m
      3. ¥i¦Û°Ê¤Æ¦a¿ëÃѨ㦳¤£¦P¸õÀW³t²v¤§¸õÀW³q«H¾÷
      4. ¥i¦Û°Ê¤Æ¦a¤Î®É¿ëÃѼħڵL½u¹q«H¸¹

      ­Y¯à¤Î®É§¹¦¨¤W­z¥D­n¥Ø¼Ð¡A«h¥i¶i¦æ¤U¦C¦¸­n¥Ø¼Ð¡G

      1. ¥i¾A¥Î©ó¤£¦P¤§°»´úÀô¹Ò¡]Âø°T¡B©wÀW°T¸¹¡B¼s¼½¹q¥xµ¥¡^
      2. ¥i¦Û°Ê¤Æ¦a¿ëÃѵL½u¹qºôÅÜ´«ÀW²v¡]©Î¸õÀWÀW²v²Õ¡^
      3. ¥i¦Û°Ê¤Æ¦a½Õ¾ã¤ÀªR³W«h¤Î¤ÀªR°Ñ¼Æ
      4. ¥i¾A¥Î©ó¦h­Ó°T¸¹ºI¦¬¾¹

      ¥»­pµe¹w­p§¹¦¨¤@®MMATLABªºµ{¦¡®w¡A±N¥]§t¶i¦æµL½u¹qºô¥x¤ÀªR©Ò¶·ªº¦UºØ¹Ï§Î¿ëÃÑ¡B¸s»E¤ÀªR¡B¤Î³n¦¡­pºâªº¦UºØ¤èªk¡C

  97. ¦³Án¹q¤l¹Ï®ÑÀ]ªº¦Û°Ê¤Æ§Þ³N»P¤u¨ã

    • ­^¤å¦WºÙ: A Study of Techniques and Tools for Audio/Textual Digital Library
    • ­pµe½s¸¹:
    • ¥D«ù¤H: ±i«T²±¡B­ð¶Ç¸q¡B±i´¼¬P
    • ¸É§U³æ¦ì: °ê¬ì·|
    • ­pµe°õ¦æ´Á¶¡: 1998/8/1 to 1999/7/1
    • ÃöÁäµü: ¹q¤l¤Æ¦³Ánµü¨å¡B¸ê°TÀ˯Á¡B¹Ï®Ñ¤ÀÃþ¡Bºô¸ô·j´M¤ÞÀº
    • ºK­n²¤¶:

      »s§@¦³Án¹q¤l¹Ï®Ñ»Ý­n¾ã¦X¤å¦r¡BÁn°T¡B»y­µ¦X¦¨µ¥µ¥¦hºØ´CÅé»P§Þ³N¡C°£¤F¦¹¤@¾ã¦X¤u§@¡A©|»Ý­n¥[¤J¤jºõ¡B¯Á¤Þµ¥µ²ºc©Êªº¸ê®Æ¡C¹ï¾ãÅ馳Án¹q¤l¹Ï®ÑÀ]¦Ó¨¥¡A§ó¥²¶·¨ã³Æ¹q¤l¤Æªº¹Ï®Ñ¥Ø¿ý¡B¥DÃD¯Á¤Þ¡B¬Æ¦Ó®Ñ¤º»P®Ñ¶¡ªº¤¬¨£¡]cross references¡^µ¥µ¥¸ê°T¡C¦p¦¹¡A¦³Án¹q¤l¹Ï®Ñ¤~¯à´£¨Ñ¤Íµ½¡B¤¬°Ê¦¡¬É­±¡C¥Ø«e³o¨Ç§@·~§¹¥þ³£¬O¤â¤u¶i¦æ¡A¤u§@·Ðº¾¡B­«½Æ¡A¤Q¤À¯Ó¶O¤H¤O¡A¦³¥²­n¬ãµo¦Û°Ê¤Æªº§Þ³N»P¤u¨ã¡C¥t¥~¬°¤F¸`¬ÙÀx¦sªÅ¶¡¡B´£°ªÁn°T«~½è¡A¬Û·í»Ý­n¿ý­µ°T¸¹ªºÀ£ÁY¡B±j¤Æ¡B®ø°£­I´ºÂø°T¡B¨M©wµoÁnªÌ¦ì¸mµ¥µ¥Án°T³B²z§Þ³N¡C

      ¹q¸£ªº¦h´CÅéÀô¹Ò¡A¤w¸g¤éÁͦ¨¼ô¡A¥i¥Hµ½¥[§Q¥Î¨Ó«Ø¥ß¦³Ánªº¹q¤l¹Ï®ÑÀ]¡A®ø°£µø»ÙªÌ±µÄ²¸ê°Tªº»Ùê¡C°ª³tªºCPU¡B±m¦â¿Ã¹õ¡B­µ®Ä¥d¡Bµøµ¡§@·~Àô¹Ò³£¤w¸g¦¨¬°¼Ð·Ç°t³Æ¡C¦h´CÅ骺³B²z§@·~¤]ºCºC§Î¦¨¤u·~¬Éªº¼Ð·Ç¡C«Ø¥ß¦³Án¹q¤l¹Ï®ÑÀ]ªº¨ä¥LªÀ·|¡B§Þ³N±ø¥ó¤]³vº¥§ïµ½¡C¨Ò¦p¡G®e©ö¨ú±o³\¦h¥Xª©«~ªº¹q¤lª©¥»¡F»y­µ§Þ³N¤Q¤À¦¨¼ô¡A¥i¥H¦b¼Ð·ÇªºPC°t³Æ¤U¡A¥Ñ¤å¦r¦X¦¨¬Û·í¦ÛµMªº»y­µ¡C

  98. ³n¦¡­pºâ¤¤ªº»~®t¹w¦ô»Pµ²ºc¿ëÃÑ

    • ­^¤å¦WºÙ: Error Estimation and Structure Identification in Soft Computing
    • ­pµe½s¸¹: NSC88-2213-E-007-007
    • ¥D«ù¤H: ±i´¼¬P
    • ¸É§U³æ¦ì: °ê¬ì·|
    • ­pµe°õ¦æ´Á¶¡: 1998/8/1 to 1999/7/1
    • ÃöÁäµü: ³n¦¡­pºâ¡]Soft Computing¡^¡B¸ê®Æ¼Ò«¬¤Æ¡]Data Modeling¡^¡BÃþ¯«¸gºô¸ô¡]Artificial Neural Networks¡^¡B¼Ò½kÅÞ¿è¡]Fuzzy Logic¡^¡B¨t²ÎÃѧO¡]System Identification¡^¡B¹Ï«¬¿ëÃÑ¡]Pattern Recognition¡^¡B¸ê®Æ±´°É¡]Data Mining¡^¡Bª¾Ãѵo±¸¡]Knowledge Discovery¡^
    • ºK­n²¤¶:

      ¥»­pµe¤D¦b©µÄò¤W¦~«×¤§°ê¬ì·|­pµe¡]"³n¦¡­pºâ¦b¸ê®Æ¼Ò«¬¤ÆªºÀ³¥Î"¡^¡A¨ä¥Øªº¦³¤T¡G

      1. ©µÄò¨ÃÀ³¥Î¤W¦~«×¤§¬ã¨s¤ß±o¡A¥H»t¯q©ó±N¤µ¦~¬ã¨s­«ÂI¶°¤¤©ó¯«¸g¼Ò½k¼Ò«¬¡]Neuro-fuzzy models¡^¤§»~®t¹w¦ô¡]Error measure estimation¡^»Pµ²ºc¿ëÃÑ¡]Structure identification¡^¡C
      2. ±N©Ò±o¤èªk»P¤ß±oÀ³¥Î©óWWW (World Wide Web)¤Wªº¹ê»Ú°ÝÃD¡A¨Ò¦puser modeling, on-line news categorization, prediction of file access intervals in a proxy serverµ¥¡A¨Ã©M¶Ç²Î¤èªk¤ñ¸û¡C
      3. Ä~ÄòºûÅ@Working Group on Data Modelingªº­º­¶¡]¦¹¬°¦bIEEE Neural Networks Council Standards Committee¤§¤UªºSubcommittee¡^¡A¨Ã¥[±j¨ä¤º®e¡C

      ³n¦¡­pºâ¡]Soft computing¡^¬Oªñ¦~¨ÓProfessor Lotfi Zadeh¡]¼Ò½kÅ޿褧¤÷¡A¥ô±Ð©óU.C. Berkeley¡^´£­Òªº·s¤è¦V¡A¨äºë¯«¤D¬Oµ²¦XÃþ¯«¸gºô¸ô¡]Artificial neural networks¡^¤Î¼Ò½kÅÞ¿è¡]Fuzzy logic¡^ªºÀuÂI¡A¨Ã»²¥H¤£¶·¾É¦¡ªº³Ì¨Î¤Æ¡]Derivative-free optimization¡^¤èªk¡A¨Ò¦p¿ò¶Çºtºâªk¡]Genetic algorithms¡^¤Î¼ÒÀÀ°h¤õªk¡]Simulated annealing¡^µ¥¡A¨Ó¹ï¸ê®Æ¤Î¬J¦³ªº±M®aª¾ÃÑ¡]Expert knowledge¡^¶i¦æ¤ÀªR¤Î·L½Õ¡]Fine tuning¡^¡A¥H«Ø¥ß¤@­Ó¨ã¦³¾Ç²ß¯à¤Oªº´¼¼z«¬¨t²Î¡A¨Ã¥i¹ï©ó®ÉÅÜ¡]Time-varying¡^ªºÀô¹Ò¶i¦æ¦Û§Ú§Y®É½Õ¾A¡]On-line adaptation¡^¡A¥HÀò¨ú³Ì¨Îµ²ªG¡C

      ¦b¼Ò«¬ªº¿ï¨ú¤W¡A³n¦¡­pºâ¬O°¾¦V©ó¨Ï¥ÎÃþ¯«¸gºô¸ô©Î¼Ò½k±Àºt¨t²Î³o¨âÃþ¼Ò«¬¡C¥Ñ©óÃþ¯«¸gºô¸ô¬O¨ã¦³¾Ç²ß©Î½Õ¾A¯à¤O¡]Learning or adaptation capability¡^ªº¶Â½c¼Ò«¬¡]Black-box model¡^¡A¦Ó¼Ò½k±Àºt¨t²Î¡]Fuzzy inference systems¡^«h¬O¯àªí¥Ü±M®aª¾ÃѪº¼Ò½k³W«h®w¨t²Î¡]Fuzzy rule-based systems¡^¡A¦]¦¹³n¦¡­pºâ¯S§O±j½Õ³o¨âªÌªºµ²¦X¡A§Î¦¨­Ý¨ã¨âªÌ¤§ªøªº¯«¸g¼Ò½k±Àºt¨t²Î¡]Neuro-fuzzy inference systems¡^¡A¨äÀ³¥Î½d³ò¬Û·í¼sªx¡AÁ|¤Z¹ï©ó¸ê®Æ©Î±M®aª¾ÃѪº¼Ò«¬¤Æ¡]Modeling¡^¡A§¡¥i¥Î±o¤W¡Cªñ´X¦~¨Ó§Ú­Ì¤w¸g¥i¥H¬Ý¨ì¨Ï¥ÎÃþ¯«¸gºô¸ô©Î¼Ò½kÅÞ¿è¡]©Î¨âªÌ­Ý³Æ¡^ªº¤p«¬®a¥Î¹q¾¹²£«~¡A¨Ò¦p¬~¦ç¾÷¡B§l¹Ð¾¹¡B¹q°Ê¨íÄG¤M¡B§N®ð¾÷¡B·Ó¬Û¾÷¡BV8Äá¿ý©ñ¼v¾÷µ¥¡C§ó¤j«¬ªºÀ³¥Î«h¥i¨£©ó¨T¨®¤ÏÂê·Ù¨®¨t²Î¡]ABS¡AAnti-lock Braking Systems¡^¤Î¶Ç°Ê¨t²Î¡]Transmission systems¡^ªº±±¨î¡A¥H¤Î¹q±è¡N¹q¨®ªº¦Û°Ê±±¨î¡CµM¦Ó¦b¹ê»ÚªºÀ³¥Î¤W¡A¤´¦³³\¦h°ÝÃD«E«Ý§JªA¡A¨Ò¦p¿é¤JÅܼƪº¿ï¨ú¡]Input selection¡^©MÅܧΡ]Transformation¡^¡B¿é¤JªÅ¶¡ªº¤À³Î¡]Input space partitioning¡^¡B¼Ò½k³W«h¼Æ¥Øªº¿ï©w¡B¯}Ãa¦¡¤Î¼Wªø¦¡ªº¾Ç²ß¡]Destructive and constructive learning¡^µ¥µ¥¡A³o¨Ç°ÝÃD³£¬O¦b¶i¦æµ²ºc¿ëÃÑ¡]Structure identification¡^®É©Ò»Ý¸Ñ¨Mªº°ÝÃD¡A¤]³£¬O¥»­pµeªº¬ã¨s­«ÂI¡C¤×¨ä­«­nªº¬O¡A§Ú­Ì§Æ±æ¯à±À¾É¥X¯à°÷§Ö³t­pºâ»~®t«ü¼Ð¡]Error measure¡^ªº¤èªk¡A¤~¯à¥¿½Tªº¤Þ¾Éµ²ºc¿ëÃѪº¶i¦æ¡C

      ¯«¸g¼Ò½k¨t²Îªº°ò¥»¾Ç²ß¤èªk¬°°f¶Ç¾Éªk¡]Back-propagation¡^¡A§Y¬°Â²³æªº±è«×¤U­°ªk¡]Gradient descent¡^©Î¬O³Ì³t¤U­°ªk¡]Steepest descent¡^¡A§ó½ÆÂø¥ý¶iªº¤èªk«h¬O¦b²Î­p©Î«D½u©Ê°jÂk¡]Nonlinear regression¡^¤¤±`¥Î¨ìªºGauss-Newton Method©Î¬OLevenberg-Marquardt Method¡C¦ý¬O³o¨Ç¤èªk³£¶·­n¥Î¨ì±è«×¦V¶q¡]Gradient vectors¡^¦Ó±è«×¦V¶q¦b½ÆÂø¨t²Î¤¤¨Ã¤£®e©ö­pºâ¡A¦]¦¹¹ï©ó¸û½ÆÂøªº¤j«¬¨t²Î¡A³n¦¡­pºâ°¾¦V©ó¨Ï¥Î¤£¶·¾É¦¡ªº³Ì¨Î¤Æ¤èªk¡]Derivative-free optimization¡^[Jang97book]¡A¨Ò¦p¿ò¶Çºtºâªk¡]Genetic algorithms¡^¡B¼ÒÀÀ°h¤õªk¡]Simulated annealing¡^¡B¤U©Y¦¡Simplex·j´M¡]Downhill simplex search¡^¡BÂø¶Ã·j´M¡]Random search¡^¡B¥¦¥¬·j´M¡]Tabu search¡^µ¥¡C³o¨Ç¤èªk¦U¦³Àu¯ÊÂI¡A¥»­pµeªº¥t¤@­«ÂI«h¦b©ó§ä¥X¦p¦ó·f°t Derivative-free ¤Î Derivative-based³o¨âºØ³Ì¨Î¤Æ¤èªk¡A¥H¨Ï¼Ò«¬¿ï¨úªº¤u§@¯à¶¶§Q¶i¦æ¡C

      ªñ¦~¨Ó¥Ñ©óºô»Úºô¸ô¡]Internet¡^¤éº¥¿³²±¡AWWW (World Wide Web)ªº¨Ï¥Î¶V¨Ó¶V´¶¹M¡A¦UºØ¹q¤l¸ê°Tªº¬y³q»P¨ú±o¤]¬O««¤â¥i±o¡C¦]¦¹¦p¦ó±q¤j¶qªº¸ê®Æ¤¤§ä¥X¦³¥Îªº¦]ªGÃö«Y¡A«K¦¨¬°¤@­Ó­«­nªº½ÒÃD¡C³o¤è­±ªº¬ã¨sºÙ¬°¸ê®Æ±´°É¡]Data Mining¡^©Îª¾Ãѵo±¸¡]Knowledge Discovery¡^¡A¤@¯ë±Ä¥Îªº¤èªk¦³²Î­p¡B²ÊÁW¶°¡]Rough Sets¡^¾÷¾¹¾Ç²ß¡]Machine Learning¡^¤¤ªºID3¡B«D°Ñ¼Æ¦¡¦^Âk¡]Nonparametric regression¡^¤¤ªºCART¡]Classification and Regression Trees¡^µ¥¡C¥»­pµe±N¹Á¸Õ¥H³n¦¡­pºâ¤Î¼Ò«¬µ²ºc¿ï¨úªº¤èªk¡A¥Î¦bWWW¤Wªº¸ê®Æ±´°É¤Îª¾Ãѵo±¸µ¥¡A¨Ò¦p User modeling, on-line news categorization ¤Î File access interval predictionµ¥¡C

      ¬°´£°ª¥xÆW¦b¦¹¤è­±¬ã¨sªºª¾¦W«×¡A¨Ã«P¶i§Þ³N¤Î¬ã¨sªº¥æ¬y¡A¥»­pµe±NÄ~ÄòºûÅ@Working Group on Data Modelingªº­º­¶¡A¦¹¬°¦bIEEE Neural Networks Council Standards Committee¤§¤UªºSubcommittee¡A URL¦a§}¬°¡G"http://www.cs.nthu.edu.tw/~jang/benchmark".

  99. µL½u¹qºô¥x¤ÀªR

    • ­^¤å¦WºÙ: The Analysis of Radio Communication Networks
    • ­pµe½s¸¹:
    • ¥D«ù¤H: ±i´¼¬P
    • ¸É§U³æ¦ì: ¤¤¬ì°|
    • ­pµe°õ¦æ´Á¶¡: 1998/8/1 to 1999/7/1
    • ÃöÁäµü: Data Clustering, Pattern Recognition, System Modeling and Simulation
    • ºK­n²¤¶:

      µL½u¹qºô¥x¤ÀªR¾¹¡]Radio Network Analyzer¡^ªº¥ô°È¬O±qºI¦¬¨ìªºµL½u¹q°T¸¹¤¤¡A³v¤@¿ëÃÑ¥X¬Û¹ï¬M¤§µL½u¹q³q«H¾÷¤§µo¸Ü¡BµL½u¹q³q«H¾÷¡A¥H¤ÎµL½u¹qºô¡C¦b¦¹´£¤Îªº¦U¶µ¦Wµü¦s¦b¤@­Ó¶¥¼h¦¡ªºÃö«Y¡G

      1. µL½u¹qºô¬O¥Ñ¼Æ­ÓµL½u¹q³q«H¾÷©Ò²Õ¦¨
      2. µL½u¹q³q«H¾÷¥i¶i¦æ¤£­­©w¦¸¼Æ¤§µo¸Ü
      3. ¨C¤@­Óµo¸Ü¬O¥Ñ¤@³s¦êªºhop©Ò²Õ¦¨

      ºI¦¬¾¹°O¿ý¤U¨ÓªºµL½u¹q°T¸¹§Y¬O¤@³s¦êªºhop¡A¦ÓµL½uºô¥x¤ÀªR¾¹ªº¥\¯à§Y¬O±NhopÂkÃþ¦Ü¥i¯à¤§µo¸Ü¡A¨Ã¶i¦Ó¤ÀªR­pºâ¥X¦¹µo¸Ü¥i¯à·½¦Û¬Y¤@µL½u¹q³q«H¾÷¡A¥H¤Î¦¹µL½u¹q³q«H¾÷¥i¯àÁõÄݪºµL½u¹qºô¡C¦]¦¹µL½u¹qºô¥x¤ÀªR¥i¤À¬°¤T¨BÆJ¡G

      1. µo¸Ü¤ÀªR¡G±Nhop ÂkÃþ¨ì¬Y¤@µo¸Ü
      2. µL½u¹q³q«H¾÷¤ÀªR¡G±Nµo¸ÜÂkÃþ¨ì¬Y¤@µL½u¹q³q«H¾÷
      3. µL½u¹qºô¤ÀªR¡G±NµL½u¹q³q«H¾÷Âk¨ì¬Y¤@µL½u¹qºô

      ¦b³o¤T­Ó¨BÆJ¤¤¡A¦³¤@­Ó¦@¦Pªº¤ÀªR¤èªk¡A´N¬O¥²¶·±N¸ê®Æ¥[¥H¤ÀÃþ¡A¦]¦¹¥»­pµeªº­º­n¥Ø¼Ð§Y¬O¹Á¸Õ¥H¦UºØ¸s»E¤ÀªR(Clustering Analysis) ¤Î¹Ï§Î¿ëÃÑ(Pattern Recognition)ªº¤èªk¡A¨Ó¶i¦æµL½u¹qºô¥x¤ÀªR¡C¶Ç²Îªº¸s»E¤ÀªR¦³¤U¦C¼ÆºØ¡G

      1. K-means clustering
      2. ISODATA
      3. Hierarchical clustering (single, complete, and average linkage)

      ¶Ç²Îªº¹Ï§Î¿ëÃѤèªk¦³¤U¦C¼ÆºØ¡G

      1. K nearest neighbor rule (KNNL)
      2. Histogram analysis
      3. Kernel and window estimators
      4. Multi-dimensional Gaussian density approach

      ³o¨Ç¶Ç²Î¤èªk³£¬O§Ú­Ì±N¹Á¸Õªº¤èªk¡C°£¦¹¤§¥~¡A§Ú­Ì¥ç¥i¥[¤J¥HÃþ¯«Ñ[ºô¸ô(Artificial Neural Networks) ©Î¼Ò½kÅÞ¿è(Fuzzy Logic)¬°°ò¦ªº¦UºØ³n¦¡­pºâ(Soft Computing) ªº¤èªk¡A¨Ò¦p:

      1. Multilayer perceptrons (MLPs)
      2. Radial basis function networks (RBFNs)
      3. Adaptive networks
      4. Fuzzy c-means clustering
      5. Adaptive fuzzy classifications
      6. Derivative-free optimization:
          Ć
        • Genetic algorithms ƒÜ
        • Simulated annealing ƒÜ
        • Random search ƒÜ
        • Downhill simplex method

      ¥»­p±Þ¹w­p§¹¦¨¤@®MMATLABªºµ{¦¡®w¡A±N¥]§t¶i¦æµL½u¹qºô¥x¤ÀªR©Ò¶·ªº¦UºØ¹Ï§Î¿ëÃÑ¡B¸s»E¤ÀªR¡B¤Î³n¦¡­pºâªº¦UºØ¤èªk¡C

  100. ºô¯¸À˯Á·j´M¤ÞÀº»Pºô¸ô§Y®É·s»DªA°È

    • ­^¤å¦WºÙ: Web Search Engines and On-line News Service and Technology
    • ­pµe½s¸¹:
    • ¥D«ù¤H: ±i´¼¬P
    • ¸É§U³æ¦ì: ¤Ó¤@«H³q
    • ­pµe°õ¦æ´Á¶¡: 1998/7/1 to
    • ÃöÁäµü: ·j´M¤ÞÀº¡Bºô¸ô¦Û°Ê¾÷¡B·s»D¦Û°Ê¤ÀÃþ¡B»EÂO¤ÀªR¡Bºô­¶¦Û°Ê¥Í¦¨¡B¤ÏÂàÀɮסB»y·N¤ÀªR
    • ºK­n²¤¶:

      ·í¨Ï¥ÎªÌ­±¹ïºô¸ô¤W»P¤é­Ñ¼Wªº¤å³¹®É¡A±`¦³¤£ª¾±q¦ó§ä°_¤§¾Ñ¡A¦pªG¯à°÷¥ý¦æ»`¶°ºô¸ô¤Wªº¦UÃþ¤å³¹¡A³B²z¦¨«K©ó¬d¸ßªº¤ÏÂàÀɮס]Inverted File¡^¡A¦A¥Ñ·j´M¤ÞÀº§Q¥Î¨Ï¥ÎªÌ´£¨Ñ¬ÛÃö¦rµü°µ¬°¬d¸ß¶µ¡A§Y¥i»´©ö¦a§ä¥X¨Ï¥ÎªÌ·Q­nªº¸ê°T¡C

      ºô¸ô§Y®É·s»DªA°Èªº¿³°_¡A¨Ï±o¤@¯ëªº¨Ï¥ÎªÌ¥i¥H»´©öªºÂsÄý¥@¬É¦U¦a¡B¦UºØ¤å¦rªº·s»Dºô¡CµM¦Ó·Q­n¾\Ū¦P¤@¨Æ¥óªº³ø¾É¡A´N¥²¶·¤£Â_ªº¤Á´«¿Ã¹õ¡A³sµ²¤£¦Pªººô¯¸¡A¬Û·í¤£«K§Q¡Cºô¸ô³t«×½wºCªºª¬ªp¤U¡A§óÄݤ£¥i¦æ¡C¦pªG¯à°÷¦Û°Ê»`¶°«Ü¦h·s»Dºô¯¸ªº¸ê®Æ¡Cºî¦X¦U­Óºô¯¸ªº·s»Dºô­¶¡A¥[¥H¤ÀÃþ¾ã²z¡A§Î¦¨¤@­Óµ²ºc¤«µM¦³§Çªº·s»D¶°ÀA¡]Web News Capsule¡^¡A´N¥i¥H¸Ñ¨M¤W­z°ÝÃD¡C

      ¥»­pµe¹w­p¬ã¨s¶}µo¤@¨t¦Cªººô¸ô¸ê·½±´°É¡B¤å¦r¤ÀªR¡Bºô­¶µ{¦¡³]­p¡A¥H¤Î¦hºØÀ˯Á§Þ¥©¡A¥H«K©ó´£¨Ñ¨C¤éºô¸ô·s»D¶°ÀA©M·j´M¤ÞÀºªº§¹µ½¸Ñ¨M¤è®×¡C

  101. ºô¸ô§Y®É·s»DªA°Èªº§Þ³N¶}µo

    • ­^¤å¦WºÙ: Web On-line News Service and Technology
    • ­pµe½s¸¹:
    • ¥D«ù¤H: ±i´¼¬P¡B±i«T²±
    • ¸É§U³æ¦ì: ¤Ó¤@«H³q
    • ­pµe°õ¦æ´Á¶¡: 1998/3/1 to 1998/7/1
    • ÃöÁäµü: ºô¸ô¦Û°Ê¾÷¡B·s»D¦Û°Ê¤ÀÃþ¡B»EÂO¤ÀªR¡Bºô­¶¦Û°Ê¥Í¦¨
    • ºK­n²¤¶:

      ºô¸ô§Y®É·s»DªA°Èªº¿³°_¡A¨Ï±o¤@¯ëªº¨Ï¥ÎªÌ¥i¥H»´©öªº¨Ó¥h¥@¬É¦U¦a¡B¦UºØ¤å¦rªº·s»Dºô¯¸¡AÂsÄý·í¤éªº·s»D¡CµM¦Ó·Q­n¾\Ū¦P¤@¨Æ¥óªº³ø¾É¡A´N¥²¶·¤£Â_ªº¤Á´«¿Ã¹õ¡A³sµ²¤£¦Pªººô¯¸¡A¬Û·í¤£«K§Q¡Cºô¸ô³t«×½wºCªºª¬ªp¤U¡A§óÄݤ£¥i¦æ¡C¦pªG¯à°÷¦Û°Ê»`¶°«Ü¦h·s»Dºô¯¸ªº¸ê®Æ¡Cºî¦X¦U­Óºô¯¸ªº·s»Dºô­¶¡A¥[¥H¤ÀÃþ¾ã²z¡A§Î¦¨¤@­Óµ²ºc¤«µM¦³§Çªº·s»D¶°ÀA¡]Web News Capsule¡^¡A´N¥i¥H¸Ñ¨M¤W­z°ÝÃD¡C¥»­pµe¹w­p¬ã¨s¶}µo¤@¨t¦Cªººô¸ô¸ê·½±´°É¡B¤å¦r¤ÀªR¡Bºô­¶µ{¦¡³]­p¡A´£¨Ñ¨C¤éºô¸ô·s»D¶°ÀAªº§¹¾ã¸Ñ¨M¤è®×¡C

  102. ²M½«¶éºô¸ô®Ñ°|µo®i­pµe

    • ­^¤å¦WºÙ: Cyber University
    • ­pµe½s¸¹:
    • ¥D«ù¤H: ¶À¤@¹A¡B·¨¨û­ë¡B±i´¼¬P
    • ¸É§U³æ¦ì: ±Ð¨|³¡
    • ­pµe°õ¦æ´Á¶¡: 1998/1/1 to 1998/7/1
    • ÃöÁäµü: WWW, Distance Learning, Cyber University
    • ºK­n²¤¶:

      ºô»Úºô¸ô(Internet)¦Û1993¦~¶}©l¿³°_¡A¦bµuµu¼Æ¦~¶¡¤£¶È¦¨¬°¥@¬É©Ê³Ì¨üÅwªïªº´CÅé¡A¦Ó¥B¤w¦¨¬°°ê»Ú¶¡¸ê°TÄvª§ªº§Q¾¹¡A¬Æ¦Ü¦¨¬°¿Å¶q¦U°ê¸ê°TÄvª§¤Oªº­«­n«ü¼Ð¡Cºô»Úºô¸ô¦b»OÆW¶}©l´¶¤Î¡A¤£¹L¬Oªñ¨â¡B¤T¦~¶¡ªº¨Æ¡A¦ý«o¤wµo®i¦¨¤Q´X¦Ü¤G¤Q´X·³«C¤Ö¦~¥Í¬¡¤¤¤é¯q­«­nªº¸ê°T´C¤¶¡A¨Ò¦p¡A¨Ì¾Ú³Ìªñ¤Ñ¤UÂø»x¤K¤Q¤»¦~¤Q¤@¤ë¥÷©Ò¶i¦æ¥xÆW²Ä¤@¦¸¥þ­±©Êªº¡uºô»Úºô¸ôªº¨Ï¥Î½Õ¬d¡vµ²ªG«ü¥X¡A¥Ø«e¤»¦¨¤»ªººô¸ô±Ú¥¼º¡¤T¤Q·³¡C¬Û¹ï¦a¡A¥Ø«e¥|¤Q·³¥H¤WªºªÀ·|¤¤°í¥÷¤l¡]¥]§t¤j¦h¼Æªº¾Ç³N¬É¤H¤h©M±Ð¨|¥DºÞ©x­û¡^¡A¹ïºô¸ô¤å¤Æ«o³»¦h¨ã¦³¤@¼Ò½kªº¦L¶H¡C¥ç§Y¡Aºô¸ô¤å¤Æªº¿³°_¤w©úÅã¦bªÀ·|¤W³y¦¨·U¨Ó·U²`ªº¥N·¾¡C

      ºô»Úºô¸ôÁöµM¨ã³Æ¦UºØ¤£¦P¥\¯à¡A¦ý¥Ø«e¨ä¦b»OÆWªºÀ³¥Î¡A«o¦h°¾­«©ó°Ó·~©Ê½è¡A©Î¶È®³¨Ó°µ¬°­Ó¤H¡B¾÷Ãö©Î¤½¥q¦æ¸¹ªº¤¶²Ð¡C³\¦h¤H¬Æ¦Ü§ó§Q¥Î¨ä¬JÁô½ª¤S¶}©ñªº¯S©Ê¡A¶}³]¨ã¦³°Ó·~¦æ¬°ªº±¡¦âºô¯¸¡A³o¨Çºô¯¸¤£¶È¼Æ¥Ø²³¦h¡A¦Ó¥B©¹©¹³£¬O¥Ø«e¤¤¤åºô¸ô¥@¬É¤¤³Ì¼öªùªº¯¸§}¡C·í§Ú­ÌªÀ·|ªººô¸ô¸ê·½¡A¦³¦p¦¹¤jªº¤ñ¨Ò³Q¥Î©ó³B²z³o¨Ç±¡¦âªº¤º®e¡A¨º»ò§Y¨Ï¬F©²§ë¸ê¦bµwÅé¤Wªº¸g¶O¦A¦h¡A³£±N¦]¦¹³Q¤j¶qµê¯Ó±¼¡C¦¹¥~¡A¤@­Ó¦³¤ß¤H¦pªG­n¹Á¸Õ¤Wºô´M³Vª¾©Êªº¤º®e®É¡A«o·|µh¤ßµo²{¥xÆWªººô¸ô¤å¤Æ³º¬O¦p¦¹³h½C¡A§Ú­Ì»Ý­nªá¶O¦n¤j¤O¶q¡A¤~§ä±o¨ì¤Ö¼Æ¤@¨Ç¸û¨ã¤º²[ªººô¯¸¡C³o¬Oªñ¦~¨Ó³Ì±`Å¥¨ìªºÁn­µ¡G¡uºô»Úºô¸ô¤W¨S¦³¤°»ò¦nªº¤¤¤åºô¯¸¡v¡A¦³¥ô±Ð¤j¾Çªº±Ð±Âµø¬°²z©Ò·íµM¦a»¡¡G¡u§Ú±q¤£¤W¤¤¤åªººô¯¸¡A¤¤¤å¯¸¨S¤°»ò¦n¬Ýªº¡A§Ú¥u¤W­^¤åªº¡v¡C

      ¦b¦¹®É¬F©²¿n·¥±À°Ê¡u¤T¦Ê¸U¤H¤Wºô¡v¡Aºô¸ô¨Ï¥Î´X¤w¦¨¬°¥þ¥Á¹B°Ê¡A§Ú­Ì¤£¸T­n°Ý¡G¡u¤¤¤åºô¸ôµo®iÀ³¦ó¥h¦ó±q¡H¡v§Ú­ÌÀ³«ä¦Ò¡G¡u¦b³oºØ¥H¦è¤è¬ì§Þ¤å©ú¥D¾Éªº¼Æ¦ìºô¸ô®ö¼é¤U¡A¤¤¤åªººô¸ô¤å¤Æ¦b¦¹·s¤@¥Nªº´CÅé­²©R¤¤À³§_¦û¦³¤@®u¦ì¸m¡H¤S¸Ó¬O½Öªº³d¥ô¡H¡v§Ú­Ì¬Û«H°ß¦³¸g¥Ñ³oºØ·NÃÑı¿ô¨Ã¹ªÀy¤j®a¿n·¥§ë¤Jºô¸ô¤¤¤å¤Æªº¦æ°Ê¡AµØ¤H¥¼¨Óªººô¸ô¥@¬É¤£­P²_¬°­^¤å¤å¤Æªº´Þ¥Á¦a¡A¤×¨ä§Ú­Ì°ª¯Åª¾ÃÑ¥÷¤l®y¾Ö°ª¬ì§ÞÀô¹Ò¡A¬Û¹ï¦a¡A¦b§Þ³N¡B¸ê·½¸ûªÀ·|¤@¯ë¤HÀu¶Õ¡A§ó¬O³dµL®Ç¶U¡A¦¹¬G¡AÁöµM§Ú­Ìªº®É¶¡©M¤O¶q³£¬Û·í¦³­­¡A¦ý§Ú­Ì¦³¤ß¥I¥X¡A´Á¬ß¯à¦h¤Ö´À³o¶ô¤g¦a¤Wªººô¸ô¤å¤Æª`¤J²M·sªº¦å²G¡C

      Äw³]²M½«¶éºô»Úºô¸ô¸ê°T¶é°Ï(Web-based Information Park)¡A¨Ã¦b¨ä¤¤¶}³]¦UºØµêÀÀ³Õª«À]¡A¦p¬ì¾ÇÀ]¡BÃÀ¤åÀ]µ¥¡A¥HÁ|¿ì¤j«¬ªºª¾©Ê³ÕÄý·|¡A¨Ã¥B³W¹ººô¸ô®Ñ°|¡A¥H®e¯Ç¦UºØ½Òµ{¡A¶i¦Ó¦³­p¹º°ö°V¦U¯Å®v¸ê¨Ã¦¨¥ß¦a°Ï©Êºô¸ô±Ð¾Ç¤¤¤ß¡Aµ²¦X¦³·NÄ@¨Ï¥Î¹q¸£°ª¬ì§Þ±Ð¾Çªº¦U¶¥¼h±Ð®v¡A´N¬O§Ú­Ì¥Ø«e§V¤Oªº¥Ø¼Ð¡C³o¼Ë¤@­Ó¼Ò¥é¯u¹ê³Õª«À]¹B§@ªº¤j«¬ºô»Úºô¸ô¸ê°T¶é°Ï¡A¦b»OÆW¡B¬Æ¦Ü¥þ¥@¬É³£¬Û·í¨u¨£¡C·í¤@¨Ç¤½¥ßªº³Õª«À]¥²¶·ªá¶O¼Æ¥H´X¤Q»õ­pªº¸g¶OÁʸm¤g¦a¨Ã«Ø¿vµwÅé®É¡A¦Ó·íºô»Úºô¸ô¥¿§Î¦¨¤@µL»·¥±©¡ªº·¾³q´C¤¶¤§»Ú¡A©Î³\³o¼Ë¤@­Ó¡u¤p¦Ó¬Ù¡v¡B¡u¤p¦Ó¦n¡v¡B¡u¥\¯à¦h´CÅé¦h¤¸¤Æ¡v¡B¡u¥ß¨¬¼Æ¦ì¦Ó©ñ²´¦a²y§ø¡vªº¥þ·sÃþ«¬ªº¶é°Ï¡A¬O­È±o§Ú­ÌªÀ·|»{¯u«ä¦Ò¨Ã§V¤Oªº¤è¦V¡C

      §Ú­Ì­p¹º¨C¦~±N´£¨Ñªñ¦Ê¶µ°ê¤º¥~¦UºØÃþ«¬ªº®iÄý¡A§Ú­Ì§Æ±æ±N¨Ó²M½«¶éºô»Úºô¸ô¸ê°T¶é¤¤ªº®iÄý¯à§óÂ×´I¦a²[»\ÃÀ¤åÃþ¡]¦p°ê¤º¥~¦UºØºô¸ôµe®i¡^¡B¬ì¾ÇÃþ¡]¦p»P¤Ñ¤å¡B¥ÍºA¡B²z¤Æ©Î¬ì§Þ¥vµ¥¤º®e¡^©M¥v¦a¥Á«UÃþµ¥¤£¦P­±¦V¡A§ó¦³¤£¦Pµ{«×ªººô®i¤º®e¡A¥H¾AÀ³¤£¦P±Ð¨|­I´ºªººô¤Íªº»Ý­n¡A¨Ã¥[¥H²Ö¿n¡A¦Ó³z¹Lºô»Ú®iÄý»s§@À窺Á|¿ì¡A§Ú­Ì¤]¦³¾÷·|±Nµo®iºë½oºô¸ô¤å¤ÆªººØ¬ó¡A´²¥¬¦ÜªÀ·|³\¦h¨¤¸¨¡C§Æ±æ¦¹¤@¸ê°T¶é°Ï¯à¦¨¬°¦UºØ¦~ÄÖ¼h©Î¤£¦P±Ð¨|µ{«×ªººô¤Í­Ì§l¨úª¾ÃÑ¡BÂX®i¨£»Dªº¤@³B¤å¤Æªø´Y¡A¨Ã´£¨ÑªÀ·|¤@²×¥Í¾Ç²ßªº­«­n´ë¹D¡C§Æ±æÂǦ¹§V¤O¶}©Ý»·¶Z¾Ç²ßªº¥t¤@­Ó·s¥@¬É¡A¸¨¹ê²×¨­±Ð¨|ªº²z·Q¡C§ó§Æ±æ¯à¦]¦¹±a°Ê¦³¤ß¤H¤hªº¦@¦P§V¤O¡AÅý¤¤¤åªººô¸ô¥@¬É¦³¾÷·|¦¨¥\¦a§êºtºô¸ô¥@¬Éª¾ÃÑ»â¯èªº¨¤¦â¡C

  103. ¤À´²¦¡¦h¦øªA¾¹ÀH·Nµø°T¨t²Î(III)(¤l­p¹º¤T) ´¼¼z«¬¬d¸ß¨t²Î (¥Hºq¿ïºq)

    • ­^¤å¦WºÙ: An Intelligent Interface of Query by Singing in VOD (Video on Deman)
    • ­pµe½s¸¹: NSC87-2213-E-007-013
    • ¥D«ù¤H: ±i´¼¬P
    • ¸É§U³æ¦ì: °ê¬ì·|
    • ­pµe°õ¦æ´Á¶¡: 1997/8/1 to 1998/7/1
    • ÃöÁäµü: Content-based Audio Retrieval, Pattern Recognition, Signal Processing
    • ºK­n²¤¶:

      ¹ï©ó¸ê®Æ®wªº¬d¸ß¤è­±¡A°£¤Fµ¹©wÃöÁä¦r¥H¶i¦æ¤@¯ëªº¬d¸ß¥~¡A§ó§Æ±æ¯à§ó¶i¤@¨B¡AÅý¨Ï¥ÎªÌ¯à¥HÁn­µÀÉ (Audio Files) ©Îµø°TÀÉ (Video Files) ªº¤º®e¨Ó½s»s¯Á¤Þ¤Î¥[³tÀ˯Á¡C¥HKTV ¬°¨Ò¡A´Á±æ¯à°µ¨ì ¡§¥Hºq¿ïºq¡¨¡A¤]´N¬O»¡¡A­nÅýºqªÌ¯à²M°Û¤@¬qºq¦±©Î±Û«ß¡A¹q¸£§Y¥H§Y®É¿ý¤U­µªi¡A¶i¦æ¥²­nªº¼Æ¬°«H¸¹³B²z¡AµM«á¤ñ¹ï¸ê®Æ®w¤¤ªº¸ê®Æ¡A¨Ì¥i¯à©Ê¦C¥X©Ò¦³¥i¯àªººq¡C°£¦¹¥H¥~¡A¤@­Ó¦ÛµMªº©µ¦ù«h¬O±NºqªÌªººqÁn¶i¦æ³B²z«á¡A§ä¥X¦U¶µ¯S©Ê¡]¦p­µ½Õ¤Î©ç¤l·Ç½T©Ê¡B­µ¦â»P­ìºqªÌ¬Û¦ü©Êµ¥¡^¨Ó°µ¦Û°Êµû¤À¡C

      ¥Ñ©ó¥»¤l­pµe²o¯A¨ìÁn­µ°T¸¹ªº³B²z¡BÃѧO¡B¤ÀÃþµ¥¡A©Ò¥H»Ý­n¦U­Ó»â°ì¯S¦³ªº§Þ³N»P¬ã¨s¬Û¤¬°t¦X¡A¥]§t¤U¦C¼Æ¶µ¡G

      1. ¼Æ¦ì°T¸¹³B²z (DSP, Digital Signal Processing): ¤×¨ä¬O¹ï©óÁn­µ°T¸¹ªºÂà´« (Transforms)»PÂoªi (Filtering)¡A¯÷¤À­z¦p¤U¡C
        • Âà´«¡G­µªi°T¸¹¥]§tªº¸ê®Æ¶q«D±`Â×´I¡A¤@¬q¤Q¬íÄÁªºÁn­µ©Ò¦ûªººÏºÐªÅ¶¡¦b¥¼À£ÁY«e¬ù¬°80 Kbyte¡]8-bit¸ÑªR«×¡A8 KHz¡^¡C±q®É°ì (Time Domain) ¤è­±¨Ó¬Ý­µªi¡A³q±`©Ò±o¦³­­¡A«ÜÃø§ä¥X©M»y­µÃѧO¬ÛÃöªº¯S¼x¶q (Features)¡C¤@¯ëªº§@ªk«h¬O±qÀW°ì (Frequency Domain) ¤è­±µÛ¤â¡A¥ç§Y¹ï­µªi¶i¦æÂ÷´²³Å¥ß¸­Âà´« (Discrete Fourier Transform)¡Aºâ¥X«e´X­Ó§CÀW¤À¶q (Low-frequency component)ªº«Y¼Æ¡A¨Ó¥Nªí©Òµ¹­µªiªº¯S©Ê¡C¨ä¥LÁÙ¦³¦UºØ¤£¦PªºÂà´«©MÅܧΧޥ©¡A¨Ò¦pWavelet Transform, Ceptral Analysis¤Î, ¥L­Ì¦U¦³¦Uªºªø³B©Mµu³B¡C
        • Âoªi (Filtering)¡G­µªiªº¨ú±o¡A¤@©w¦h¤Ö·|³Q¤£¬ÛÃöªº°T¸¹¡]§Y¾¸­µ¡^©Ò¦Ã¬V¡C¦]¦¹¦b¶i¦æ­µªiªº³B²z¤§«e¡A¤@©w­n¶i¦æÂoªi¡CÂoªiªº¤è¦¡¦³«Ü¦hºØ¡A³Ì²³æªº¤è¦¡¬O±N°T¸¹³q¹L¤@­Ó§C³qÂoªi¾¹ (Low-pass Filter)¡C§ó½ÆÂøªº¤èªk«h¬O±N¦¹Âoªi¾¹ªº¯S©ÊÅܬ°§Y®É¥i½Õ (On-line Adaptive) ¡C
      2. ¹Ï§ÎÃѧO (Pattern Recognition): ¤×¨ä¬O»y­µÃѧO (Speech Recognition) ¤Î»yªÌÃѧO (Speaker Recognition)¡CÁn­µ°T¸¹ªº¶q³q±`«Ü¤j¡A©Ò¥H¦b¶i¦æDSP¥H§ì¨ú¯S¼x¶q«á¡A¤´»Ý¶i¦æ¸ê®ÆªºÁY´î (Data Reduction)¡A¥H§Q¹Ï§ÎÃѧOªº¶i¦æ¡C¦b³o¤è­±³Ì±`¥Î¨ìªº¤èªk¬OCondense¤ÎEditing¡C¹ï©ó³o¨âºØ¤èªk¦b»yªÌ¿ë»{¤è­±ªºÀ³¥Î¡A§Ú­Ì´¿¸g¥[¥H§ï¨}¡AÀò­P¤£¿ùªºµ²ªG¡C
      3. ³n¦¡­pºâ (Soft Computing) : ¥]§tÃþ¯«¸gºô¸ô (Artificial Neural Networks)¡B¼Ò½kÅÞ¿è (Fuzzy Logic) ¡A¥H¤Î¦UºØµL¶·¾É¦¡ (Derivative-free) ªº³Ì¨Î¤Æ (Optimization) ¤èªk¡C¦b¹ê§@¤W¡A§Ú­Ì±`±`»Ý­n§ä¥X¤@­Óµ¹©wªº¥Ø¼Ð¨ç¼Æªº³Ì¤p­È¡A¦¹¥Ø¼Ð¨çÃD¥i¯à¤w¬Û·íÁc½Æ¡A¨ä±è«× (Gradient Vector) ¥i¯àÃø¥H­pºâ¡A¦P¦¹§Ú­ÌµLªk¥H¶Ç²Î³Ì¨Î¤Æªº¤èªk¨Ó¨D¸Ñ¡C¸Ñ¨M¤§¹D¤D¿ï¥Î³n¦¡­pºâ (Soft Comping) ¤¤ªº¦UºØµL¶·¾É¦¡ (Derivative-free) ªº¤èªk¡A¨Ò¦p°ò¦]ºtºâªk (Genetic Algorithms)¡A¼ÒÀÀ°h¤õªk (Simulated Annealing)¡AÂø¶Ã·j´Mªk (Random Search Method)¡A¥H¤Î¤U©YSimplexªk (Downhill Simplex Method)¡C

  104. ³n¦¡­pºâ¦b¸ê®Æ¼Ò«¬¤ÆªºÀ³¥Î

    • ­^¤å¦WºÙ: Soft Computing in Data Modeling
    • ­pµe½s¸¹: NSC87-2213-E-007-009
    • ¥D«ù¤H: ±i´¼¬P
    • ¸É§U³æ¦ì: °ê¬ì·|
    • ­pµe°õ¦æ´Á¶¡: 1997/8/1 to 1998/7/1
    • ÃöÁäµü: ³n¦¡­pºâ¡]Soft Computing¡^¡B¸ê®Æ¼Ò«¬¤Æ¡]Data Modeling¡^¡BÃþ¯«¸gºô¸ô¡]Artificial Neural Networks¡^¡B¼Ò½kÅÞ¿è¡]Fuzzy Logic¡^¡B¨t²ÎÃѧO¡]System Identification¡^¡B¹Ï«¬¿ëÃÑ¡]Pattern Recognition¡^¡B¸ê®Æ±´°É¡]Data Mining¡^¡Bª¾Ãѵo±¸¡]Knowledge Discovery¡^
    • ºK­n²¤¶:

      ³n¦¡­pºâ¡]Soft Computing¡^¬Oªñ¦~¨ÓProfessor Lotfi Zadeh¡]¼Ò½kÅ޿褧¤÷¡A¥ô±Ð©óU.C. Berkeley¡^´£­Òªº·s¤è¦V¡A¨äºë¯«¤D¬Oµ²¦XÃþ¯«¸gºô¸ô¡]Artificial Neural Networks¡^¤Î¼Ò½kÅÞ¿è¡]Fuzzy Logic¡^ªºÀuÂI¡A¨Ã»²¥H¤£¶·¾É¦¡ªº³Ì¨Î¤Æ¡]Derivative free Optimization¡^¤èªk¡A¨Ò¦p¿ò¶Çºtºâªk¡]Genetic Algorithms¡^¤Î¼ÒÀÀ°h¤õªk¡]Simulated Annealing¡^µ¥¡A¨Ó¹ï¸ê®Æ¤Î¬J¦³ªº±M®aª¾ÃÑ¡]Expert Knowledge¡^¶i¦æ¤ÀªR¤Î·L½Õ¡]Fine tuning¡^¡A¥H«Ø¥ß¤@­Ó¨ã¦³¾Ç²ß¯à¤Oªº´¼¼z«¬¨t²Î¡A¨Ã¥i¹ï©ó®ÉÅÜ¡]Time-varying¡^ªºÀô¹Ò¶i¦æ¦Û§Ú§Y®É½Õ¾A¡]On-line Adaptation¡^¡A¥HÀò¨ú³Ì¨Îµ²ªG¡C

      ¦b¼Ò«¬ªº¿ï¨ú¤W¡A³n¦¡­pºâ¬O°¾¦V©ó¨Ï¥ÎÃþ¯«¸gºô¸ô¦¨¼Ò½kÅÞ¿è³o¨âÃþ¼Ò«¬¡C¥Ñ©óÃþ¯«¸gºô¸ô¬O¨ã¦³¾Ç²ß©Î½Õ¾A¯à¤O¡]Learning or Adaptation Capability¡^ªº¶Â½c¼Ò«¬¡]Blackbox Model¡^¡A¦Ó¼Ò½k±Àºt¨t²Î¡]Fuzzy Inference Systems¡^«h¬O¯àªí¥Ü±M®aª¾ÃѪº¼Ò½k³W«h®w¨t²Î¡]Fuzzy Rule-based Systems¡^¡A¦]¦¹³n¦¡­pºâ¯S§O±j½Õ³o¨âªÌªºµ²¦X¡A§Î¦¨­Ý¨ã¨âªÌ¤§ªøªº¯«¸g¼Ò½k±Àºt¨t²Î¡]Neuro-Fuzzy Inference Systems¡^¡A¨äÀ³¥Î½d³ò¬Û·í¼sªx¡AÁ|¤Z¹ï©ó¸ê®Æ©Î±M®aª¾ÃѪº¼Ò«¬¤Æ¡]Modeling¡^¡A§¡¥i¥Î±o¤W¡Cªñ´X¦~¨Ó§Ú­Ì¤w¸g¥i¥H¬Ý¨ì¨Ï¥ÎÃþ¯«¸gºô¸ô©Î¼Ò½kÅÞ¿è¡]©Î¨âªÌ­Ý³Æ¡^ªº¤p«¬®a¥Î¹q¾¹²£«~¡A¨Ò¦p¬~¦ç¾÷¡B§l¹Ð¾¹¡B¹q°Ê¨íÄG¤M¡B§N®ð¾÷¡B·Ó¬Û¾÷¡BV8Äá¿ý©ñ¼v¾÷µ¥¡C§ó¤j«¬ªºÀ³¥Î«h¥i¨£©ó¨T¨®¤ÏÂê·Ù¨®¨t²Î¡]ABS¡AAnti-lock Braking Systems¡^¤Î¶Ç°Ê¨t²Î¡]Transmission Systems¡^ªº±±¨î¡A¥H¤Î¹q±è¡N¹q¨®ªº¦Û°Ê±±¨î¡CµM¦Ó¦b¹ê»ÚªºÀ³¥Î¤W¡A¤´¦³³\¦h°ÝÃD«E«Ý§JªA¡A¨Ò¦p¿é¤JÅܼƪº¿ï¨ú¡]Input Selection¡^©MÅܧΡ]Transformation¡^¡B¿é¤JªÅ¶¡ªº¤À³Î¡]Input Space Partitioning¡^¡B¼Ò½k³W«h¼Æ¡]Number of Fuzzy Rules¡^ªº¿ï©w¡B¯}Ãa¦¡¤Î¼Wªø¦¡ªº¾Ç²ß¡]Destructive and Constructive Learning¡^µ¥µ¥¡A³o¨Ç³£¬O¥»­pµeªº¬ã¨s­«ÂI¡C

      ¯«¸g¼Ò½k¨t²Îªº°ò¥»¾Ç²ß¤èªk¬°°f¶Ç¾Éªk¡]Backpropagation¡^¡A§Y¬°Â²³æªº±è«×¤U­°ªk¡]Gradient Descent¡^©Î¬O³Ì³t¤U­°ªk¡]Steepest Descent¡^¡A§ó½ÆÂø¥ý¶iªº¤èªk«h¬O¦b²Î­p©Î«D½u©Ê°jÂk¡]Nonlinear Regression¡^¤¤±`¥Î¨ìªºGauss-Newton Method©Î¬OLevenberg-Marquardt Method¡C¦ý¬O³o¨Ç¤èªk³£¶·­n¥Î¨ì±è«×¦V¶q¡]Gradient Vector¡^¦Ó±è«×¦V¶q¦b½ÆÂø¨t²Î¤¤¨Ã¤£®e©ö­pºâ¡A¦]¦¹¹ï©ó¸û½ÆÂøªº¤j«¬¨t²Î¡A³n¦¡­pºâ°¾¦V©ó¨Ï¥Î¤£¶·¾É¦¡ªº³Ì¨Î¤Æ¤èªk¡]Derivative-free Optimization¡^¡A¨Ò¦p¿ò¶Çºtºâªk¡]Genetic Algorithms¡^¡B¼ÒÀÀ°h¤õªk¡]Simulated Annealing¡^¡B¤U©Y¦¡Simplex·j´M¡]Downhill Simplex Search¡^¡BÂø¶Ã·j´M¡]Random Search¡^¡B¥¦¥¬·j´M¡]Tabu Search¡^µ¥¡C³o¨Ç¤èªk¦U¦³Àu¯ÊÂI¡A¥»­pµeªº¥t¤@­«ÂI«h¦b©ó§ä¥X¦pªG¿ï¨ú³o¨Ç¤èªkªº±±¨î°Ñ¼Æ¡]Control Parameters¡^ªº¨BÆJ¡A¥H«K¨Ï¥Î©ó¸ê®Æ¼Ò«¬¤Æ¡]Data Modeling¡^¤§¤W¡C

      ªñ¦~¨Ó¥Ñ©óºô»Úºô¸ô¡]Internet¡^¤éº¥¿³²±¡AWWW (World Wide Web)ªº¨Ï¥Î¶V¨Ó¶V´¶¹M¡A¦UºØ¹q¤l¸ê°Tªº¬y³q»P¨ú±o¤]¬O««¤â¥i±o¡C¦]¦¹¦p¦ó±q¤j¶qªº¸ê®Æ¤¤§ä¥X¦³¥Îªº¦]ªGÃö«Y¡A«K¦¨¬°¤@­Ó­«­nªº½ÒÃD¡C³o¤è­±ªº¬ã¨sºÙ¬°¸ê®Æ±´°É¡]Data Mining¡^©Îª¾Ãѵo±¸¡]Knowledge Discovery¡^¡A¤@¯ë±Ä¥Îªº¤èªk¦³²Î­p¡B²ÊÁW¶°¡]Rough Sets¡^¾÷¾¹¾Ç²ß¡]Machine Learning¡^¤¤ªºID3¡B«D°Ñ¼Æ¦¡¦^Âk¡]Nonparametric Regression¡^¤¤ªºCART¡]Classification and Regression Trees¡^µ¥¡C¥»­pµe±N¹Á¸Õ¥H³n¦¡­pºâ¥Î¦b¸ê®Æ¼Ò«¬¤Æªº§Þ¥©¡A¥Î¦b¸ê®Æ±´°É¤Îª¾Ãѵo±¸ªº¦UºØBenchmark Problems¡C

  105. »yªÌ¿ë»{

    • ­^¤å¦WºÙ: Speaker Recognition
    • ­pµe½s¸¹: NSC 86-2213-E-007-048
    • ¥D«ù¤H: ±i´¼¬P
    • ¸É§U³æ¦ì: °ê¬ì·|
    • ­pµe°õ¦æ´Á¶¡: 1996/8/1 to 1997/7/31
    • ÃöÁäµü: Speaker recognition, pattern recognition, neuro-fuzzy modeling, artificial neural networks, fuzzy logic, digital signal processing
    • ºK­n²¤¶:
      With the advance of modern high-speed computers, now we can try computation intensive approaches that were deemed too inefficient for practical problems. These approaches include adaptive learning systems such as artificial neural networks and adaptive networks, and innovative optimization techniques such as genetic algorithms (GA) and simulated annealing. These approaches, together with fuzzy set theory as a knowledge representation tool, form the constituents of the so-called soft computing that has been used for real-world problems such as character recognition, color recipe prediction and adaptive control.

      This project applies the aforementioned soft computing techniques to a challenging real-world problem: automatic speaker recognition (ASR). Given a speech input, the objective of ASR is to output the identity of the person most likely to have spoken. One application of ASR is to enhance human-machine interface. For instance, voice activated computer should be programmed to adapt and respond to the current user. Security applications of ASR are plenty, for instance, security check when entering a building or accessing a bank account. Moreover, ASR has the convenience of easy data collection over the telephone.

      This project emphasizes on both research and software/hardware implementation. ASR is a difficult problem in pattern recognition. It involves typically a huge amount of data and we need to apply digital signal processing techniques to down-size the data dimension and extract relevant features for further processing of data classification or discriminant analysis. For such a difficult problem, a single approach is usually not enough and we need a collection of various methodologies to complement each other to accomplish the task.

      For research part, we will tackle ASR with both soft-computing techniques and conventional statistical pattern recognition. We have been working on neuro-fuzzy and soft-computing techniques for several years and the applications include time series prediction, data classification, nonlinear system identification, noise cancellation, channel equalization, adaptive control, printed character recognition, and inverse kinematics problems. We shall apply the soft-computing techniques (neural networks, fuzzy logic, adaptive neuro-fuzzy systems, genetic algorithms and simulated annealing) we gained over years to ASR, and complement it with conventional statistical pattern recognition such as Baysian approach.

      For software implementation, our primary tools are MATLAB and C. MATLAB is an integrating environment for scientific computation and data visualization tool. We have positive experiences using MATLAB to deliver GUI-based fuzzy product [], and we expect to have GUI based demo as the product of this project. For computation-intensive and non-vectorizable operation, we will resort to C language for high speed.

      For hardware implementation, our goal is to set up a hardware system using a Pentium PC and dSPACE 1102 controller board to take audio signal from a speaker, do FFT and feature extraction, feed the features to a trained classifier, and return the identity of the speaker on the fly. The whole process is time consuming; it is virtually impossible to do on-line identification without hardware support.

      To sum up, this project is well balanced in terms of research and implementation. We will benefit from the research of using soft computing and statistical approaches for speaker recognition; this paves the avenue to a more difficult problem of speech recognition. The hardware implementation can prove its feasibility and provide a demonstration for further exploration and possible commercialization.