Abstract
![CDATA[Speech register research shows that humans are adept at finetuning components of their speech to accommodate the needs audience of the audience, suggesting that they have a model of human communication needs. However, when that audience is a computer rather than another human, such a model may be invalid. Here we examine humans’ speech to other humans or an auditory-visual avatar before and after the computer or the human listener makes a listening “error”. Speech is found to be hyperarticulated in Computer- compared with Human- Directed speech, and also in speech after correction. Results are discussed in terms of human-computer interaction and ASR systems.]]
Original language | English |
---|---|
Title of host publication | Proceedings of the 13th Australasian International Conference on Speech Science and Technology, 14-16 December 2010, Melbourne, Australia |
Publisher | ASSTA |
Pages | 13-17 |
Number of pages | 5 |
ISBN (Print) | 9780958194631 |
Publication status | Published - 2010 |
Event | Australasian International Conference on Speech Science and Technology - Duration: 3 Dec 2012 → … |
Conference
Conference | Australasian International Conference on Speech Science and Technology |
---|---|
Period | 3/12/12 → … |
Keywords
- human-computer interaction
- speech