17-3 Digit Recognition: Varying MFCC Dimensions (?¸å?辨è?:改變MFCC維度)

¦b¤W¤@¸`¤¤¡A§Ú­Ì¤w¸g¨Ï¥Î¤¤¤å¼Æ¦r¿ëÃѨӻ¡©ú¤F HTK ªº°ò¥»¨Ï¥Î¤è¦¡¡A¦b¥»¸`¤Î¥H¤U¦U¸`¤¤¡A§Ú­Ì±N»¡©ú¦p¦ó§ïÅܦUºØ°Ñ¼Æ¡]¥]§t»y­µ¯S¼x¡BÁn¾Ç¼Ò«¬¬[ºcµ¥¡^¡A¥H«K§ï¶i¿ëÃѲv¡C

­º¥ý¡A§Ú­Ì±N°ò¥»°V½m¤Î´ú¸Õªºµ{¦¡¥]¦¨¤@­Ó¨ç¼Æ htkTrainTest.m¡A¥H«K¯à°÷¤ÏÂЩI¥s¡A¦¹¨ç¼Æ±µ¦¬¤@­Óµ²ºcÅÜ¼Æ htkPrm¡A¥]§t°V½m©Ò¥Î¨ìªº¦UºØ°Ñ¼Æ¡C

­º¥ý¡A¦bÁn¾Ç¼Ò«¬¬[ºc¤£Åܪº±¡ªp¤U¡A§Ú­Ì¥i¥H§ïÅÜ»y­µ¯S¼x¡A¤W¤@¸`½d¨Ò©Ò¥Îªº»y­µ¯S¼x¬O 13 ºûªº MFCC_E¡]MFCC & Energy¡^¡A§Ú­Ì¥i¥H¥[¨ì 26 ºûªº MFCC_E_D ©Î¬O MFCC_E_D_Z¡A¤]¥i¥H¥[¨ì 39 ºûªº MFCC_E_D_A ©Î MFCC_E_D_A_Z¡C¨ä¤¤

­Y¨Ï¥Î 26 ºûªº MFCC_E_D_Z¡A¥i¨£¤U¦C½d¨Ò¡G

Example 1: htk/chineseDigitRecog/training/goSyl26.mhtkPrm=htkParamSet; htkPrm.pamFile='digitSyl.pam'; htkPrm.feaCfgFile='mfcc26.cfg'; htkPrm.feaType='MFCC_E_D_Z'; htkPrm.feaDim=26; htkPrm.streamWidth=[26]; disp(htkPrm) [trainRR, testRR]=htkTrainTest(htkPrm); fprintf('Inside test = %g%%, outside test = %g%%\n', trainRR, testRR); pamFile: 'digitSyl.pam' feaCfgFile: 'mfcc26.cfg' waveDir: '..\waveFile' sylMlfFile: 'digitSyl.mlf' phoneMlfFile: 'digitSylPhone.mlf' mnlFile: 'digitSyl.mnl' grammarFile: 'digit.grammar' feaType: 'MFCC_E_D_Z' feaDim: 26 mixtureNum: 3 stateNum: 3 streamWidth: 26 Pruning-Off Pruning-Off Pruning-Off Pruning-Off Pruning-Off Inside test = 91.29%, outside test = 92.86%

¹ïÀ³ªº batch ÀɮסA½Ð¨£ goSyl26.bat¡C

­Y¨Ï¥Î 39 ºûªº MFCC_E_D_A_Z¡A½Ð¨£¤U¦C½d¨Ò¡G

Example 2: htk/chineseDigitRecog/training/goSyl39.mhtkPrm=htkParamSet; htkPrm.pamFile='digitSyl.pam'; htkPrm.feaCfgFile='mfcc39.cfg'; htkPrm.feaType='MFCC_E_D_A_Z'; htkPrm.feaDim=39; htkPrm.streamWidth=[39]; disp(htkPrm) [trainRR, testRR]=htkTrainTest(htkPrm); fprintf('Inside test = %g%%, outside test = %g%%\n', trainRR, testRR); pamFile: 'digitSyl.pam' feaCfgFile: 'mfcc39.cfg' waveDir: '..\waveFile' sylMlfFile: 'digitSyl.mlf' phoneMlfFile: 'digitSylPhone.mlf' mnlFile: 'digitSyl.mnl' grammarFile: 'digit.grammar' feaType: 'MFCC_E_D_A_Z' feaDim: 39 mixtureNum: 3 stateNum: 3 streamWidth: 39 Pruning-Off Pruning-Off Pruning-Off Pruning-Off Pruning-Off Inside test = 91.07%, outside test = 92.86%

¹ïÀ³ªº batch ÀɮסA½Ð¨£ goSyl39.bat¡C

¦b¹ïÀ³ªº batch Àɮפ譱¡A§Ú­Ì¨Ã¨S¦³¥]¦¨¨ç¼Æ¨Ó°õ¦æ¡A©Ò¥H¨ä¤º®e¬Ý°_¨Ó·|¤ñ¸ûÁc½Æ¡C±q goSyl13.bat ºtÅܨì goSyl26.bat¡A¨Æ¹ê¤W¥u§ïÅܤF¨â¦C¡AŪªÌ¥i¥H¨Ï¥Î¤U¦C«ü¥O¨Ó¤ñ¸û¨ä®t²§¡G

fc goSyl13.bat goSyl26.bat
§Q¥Î¬Û¦Pªº¤èªk¡A¤]¥i¥H¤ñ¸û goSyl26.bat ©M goSyl39.bat ªº®t²§¡C
Audio Signal Processing and Recognition (­µ°T³B²z»P¿ëÃÑ)