Report on Session I: Prosodic Aids to Speech Recognition
نویسنده
چکیده
Four papers were presented in the opening session of the conference. The papers were "Prosody and Parsing" by P. Pennsylvania). Price et al. reported on the use of prosodic information to resolve several types of syntactic ambiguities, the development of a prosodic information coding system suitable for a parser, and the development of automatic algorithms for extracting prosodic information. (Work jointly supported by NSF and DARPA.) Mary Beckman reported on work in articulatory dynamics which suggests a new approach to the use of durational information in continuous speech recognition. New models of articulatory gesture allow for useful distinctions among the timing effects found in global tempo increase, phrase-final lengthening, and sentence accent. (Work supported by NSF.) Julia Hirschberg reported on work in empirical observation of the pragmatic uses of selected pitch contours. In addition, her report addressed the need for better speech data (goal-directed speech in a specific task domain) on which to test hypotheses about the interaction of prosodic constructs with the other components of a spoken language understanding system, particularly semantics and pragmatics. (Work supported by AT~T Bell Laboratories.) Mark Steedman reported on work in the description of intonational and syntactic structures in a combinatory extension of categorlal grammar. Combinatory categorial grammar predicts syntactic units which align with boundaries in the intonational structure, thus helping to clarify the structure of an utterance for spoken language understanding.
منابع مشابه
Prosodic elements to improve pronunciation in English language learners: A short report
The usefulness of teaching pronunciation in language instruction remains controversial. Though past research suggests that teachers can make little or no difference in improving their students’ pronunciation, current findings suggest that second language pronunciation can improve to be near native-like with the implementation of certain criteria such as the utilization of...
متن کاملA Database for Automatic Persian Speech Emotion Recognition: Collection, Processing and Evaluation
Abstract Recent developments in robotics automation have motivated researchers to improve the efficiency of interactive systems by making a natural man-machine interaction. Since speech is the most popular method of communication, recognizing human emotions from speech signal becomes a challenging research topic known as Speech Emotion Recognition (SER). In this study, we propose a Persian em...
متن کاملAlignment of human prosodic patterns for spoken dialogue systems
An adaptive speech recognizer is a key function in the design of a robust spoken dialogue system. Our research focuses on the human tendency of prosodic alignment to one’s conversational partners. A spoken dialogue system might be able to exploit this human tendency to implicitly influence people to manage their speech at the prosodic level in order to accommodate its recognition capabilities. ...
متن کاملToBI Prosodic Analysis of a Professional Speaker of American English
We analyze the distribution of ToBI labels in a corpus collected from a professional speaker for use in concatenative speech synthesis. Our goals include using such statistics to aid automatic ToBI labeling of such a corpus, analogously to how a language model aids speech recognition. We find that the professional speaker produces a rich variety of prosodic events. ToBI labels occur with skewed...
متن کاملذخیره در منابع من
با ذخیره ی این منبع در منابع من، دسترسی به آن را برای استفاده های بعدی آسان تر کنید
عنوان ژورنال:
دوره شماره
صفحات -
تاریخ انتشار 1989