Overview

A one-day workshop on Human and Machine Speech will be hosted by HCSNet with ASSTA involvement. The workshop will be held on Friday December 1, 2006 at Sancta Sophia College within the grounds of the University of Sydney.

In this event the likely synergies between Human and Machine speech recognition will be explored. Specifically, it seems plausible that understanding human speech recognition will provide insights into how automatic speech recognition (ASR) might be improved. For example, it is possible that properties that facilitate human speech perception might also assist in developing ASR training and test regimes. Furthermore, studies in human speech perception and development have suggested that there are other aspects inherent in different types of speech signal (such as affective and musical qualities) that overlap with language learning. Thus the design of ASR systems might benefit from a more expansive conception of speech phenomena. On the other hand, we know the conditions under which humans learn speech, but not exactly how they come to do it. Here, insights gained from implementing machine systems have the potential to throw light on constraints inherent in the development of human speech perception. Indeed, computational modeling of speech perception and development provide a test-bed in which models of Human Speech Recognition (HSR) can be evaluated. Furthermore, a consideration of constraints on aspects of communication brings into play disciplines as diverse as linguistics, communication disorders, music cognition and philosophy and investigators from all these areas should find something to interest them in this workshop.

The goals of the Human and Machine Speech workshop are:

• To bring together researchers interested in Human speech processing and development, multimodal speech processing, speech coding and ASR systems;
• To encourage interactions and collaboration within this community and within the wider international speech community;
• To foster interaction between academic and industrial researchers;
• To encourage dissemination of research results;
• To provide a forum for the discussion of new and ongoing research and projects;
• To provide an opportunity for the broader artificial intelligence community to become aware of local research; and, finally,
• To increase visibility of speech research in Australia, New Zealand and overseas.

Workshop Program

A program for the workshop is now available by clicking here.

Workshop Organisers

Associate Professor Christopher Davis
Professor Denis Burnham
Associate Professor Jeesun Kim

Enquiries

For any comments or questions about these pages please contact the organisers.

Attachment	Size
H&M Speech Program.pdf	64.58 KB