Spoken Language Understanding Systems 1
HomeFull List of Titles
1: ICSLP'98 Proceedings
Keynote SpeechesText-To-Speech Synthesis 1Spoken Language Models and Dialog 1Prosody and Emotion 1Hidden Markov Model Techniques 1Speaker and Language Recognition 1Multimodal Spoken Language Processing 1Isolated Word RecognitionRobust Speech Processing in Adverse Environments 1Spoken Language Models and Dialog 2Articulatory Modelling 1Talking to Infants, Pets and LoversRobust Speech Processing in Adverse Environments 2Spoken Language Models and Dialog 3Speech Coding 1Articulatory Modelling 2Prosody and Emotion 2Neural Networks, Fuzzy and Evolutionary Methods 1Utterance Verification and Word Spotting 1 / Speaker Adaptation 1Text-To-Speech Synthesis 2Spoken Language Models and Dialog 4Human Speech Perception 1Robust Speech Processing in Adverse Environments 3Speech and Hearing Disorders 1Prosody and Emotion 3Spoken Language Understanding Systems 1Signal Processing and Speech Analysis 1Spoken Language Generation and Translation 1Spoken Language Models and Dialog 5Segmentation, Labelling and Speech Corpora 1Multimodal Spoken Language Processing 2Prosody and Emotion 4Neural Networks, Fuzzy and Evolutionary Methods 2Large Vocabulary Continuous Speech Recognition 1Speaker and Language Recognition 2Signal Processing and Speech Analysis 2Prosody and Emotion 5Robust Speech Processing in Adverse Environments 4Segmentation, Labelling and Speech Corpora 2Speech Technology Applications and Human-Machine Interface 1Large Vocabulary Continuous Speech Recognition 2Text-To-Speech Synthesis 3Language Acquisition 1Acoustic Phonetics 1Speaker Adaptation 2Speech Coding 2Hidden Markov Model Techniques 2Multilingual Perception and Recognition 1Large Vocabulary Continuous Speech Recognition 3Articulatory Modelling 3Language Acquisition 2Speaker and Language Recognition 3Text-To-Speech Synthesis 4Spoken Language Understanding Systems 4Human Speech Perception 2Large Vocabulary Continuous Speech Recognition 4Spoken Language Understanding Systems 2Signal Processing and Speech Analysis 3Human Speech Perception 3Speaker Adaptation 3Spoken Language Understanding Systems 3Multimodal Spoken Language Processing 3Acoustic Phonetics 2Large Vocabulary Continuous Speech Recognition 5Speech Coding 3Language Acquisition 3 / Multilingual Perception and Recognition 2Segmentation, Labelling and Speech Corpora 3Text-To-Speech Synthesis 5Spoken Language Generation and Translation 2Human Speech Perception 4Robust Speech Processing in Adverse Environments 5Text-To-Speech Synthesis 6Speech Technology Applications and Human-Machine Interface 2Prosody and Emotion 6Hidden Markov Model Techniques 3Speech and Hearing Disorders 2 / Speech Processing for the Speech and Hearing Impaired 1Human Speech ProductionSegmentation, Labelling and Speech Corpora 4Speaker and Language Recognition 4Speech Technology Applications and Human-Machine Interface 3Utterance Verification and Word Spotting 2Large Vocabulary Continuous Speech Recognition 6Neural Networks, Fuzzy and Evolutionary Methods 3Speech Processing for the Speech-Impaired and Hearing-Impaired 2Prosody and Emotion 72: SST Student Day
SST Student Day - Poster Session 1SST Student Day - Poster Session 2Author Index
A B C D E F G H I
J K L M N O P Q R
S T U V W X Y Z
Multimedia Files
GALAXY-II: A Reference Architecture for Conversational System Development
Authors:
Stephanie Seneff, MIT Laboratory for Computer Science (USA)
Ed Hurley, MIT Laboratory for Computer Science (USA)
Raymond Lau, MIT Laboratory for Computer Science (USA)
Christine Pao, MIT Laboratory for Computer Science (USA)
Philipp Schmid, MIT Laboratory for Computer Science (USA)
Victor Zue, MIT Laboratory for Computer Science (USA)
Page (NA) Paper number 1153
Abstract:
GALAXY is a client-server architecture for accessing on-line information using spoken dialogue which was first introduced at ICSLP-94. It has served as the testbed for developing human language technologies for our group for several years. Recently, we have initiated a significant redesign of its architecture to enable many researchers to develop their own applications, using either exclusively their own servers or intermixing them with servers developed by others. This redesign was done in part due to the fact that GALAXY has been designated as the prototype reference architecture for the new DARPA Communicator Program. The new architecture, GALAXY-II, makes use of a scripting language for flow control to provide flexible interaction among the servers, and a set of libraries to support rapid prototyping of new servers. In this paper, we describe the new architecture in some detail, and report on the current status of its development.
SL981153.PDF (From Author)
SL981153.PDF (Rasterized)
TOP
Improvements in Speech Understanding Accuracy Through the Integration of Hierarchical Linguistic, Prosodic, And Phonological Constraints in the Jupiter Domain
Authors:
Grace Chung, MIT Laboratory of Computer Science (USA)
Stephanie Seneff, MIT Laboratory of Computer Science (USA)
Page (NA) Paper number 603
Abstract:
This paper explores some issues in designing conversational systems with integrated higher level constraints. We experiment with a configuration that combines a context-dependent acoustic front-end, using MIT's SUMMIT recognizer, with ANGIE, a hierarchical framework that models word substructure and phonological processes, and with TINA, a trainable probabilistic natural language (NL) model. Working in the Jupiter weather domain, we develop a computationally tractable system which incorporates higher level linguistic, prosodic and phonological constraints together in the second of a two-pass strategy. Experiments are evaluated using a new understanding performance metric, and the new integrated system achieves up to 17.1% relative reduction in understanding error and 15.4% reduction in word error. In addition, we investigate the possibilities of a two-pass system which relies on the first stage for pruning based on syllable-level constraint, and applies linguistic and prosodic knowledge largely at the second stage.
SL980603.PDF (From Author)
SL980603.PDF (Rasterized)
TOP
Towards Robust Methods for Spoken Document Retrieval
Authors:
Kenney Ng, MIT Laboratory for Computer Science (USA)
Page (NA) Paper number 1088
Abstract:
In this paper, we investigate a number of robust indexing and retrieval methods in an effort to improve spoken document retrieval performance in the presence of speech recognition errors. In particular, we examine expanding the original query representation to include confusible terms; developing a new document-query retrieval measure based on approximate matching that is less sensitive to recognition errors; expanding the document representation to include multiple recognition hypotheses; modifying the original query using automatic relevance feedback to include new terms found in the top ranked documents; and combining information from multiple subword unit representations. We study the different methods individually and then explore the effects of combining them. Experiments on radio broadcast news data show that using a combination of these methods can improve retrieval performance by over 20%.