University of Cambridge > Talks.cam > Inference Group > Getting it Right the Second Time: Recognition of Spoken Corrections

Getting it Right the Second Time: Recognition of Spoken Corrections

Add to your list(s) Download to your calendar using vCal

If you have a question about this talk, please contact Ryan Prescott Adams.

In previous work, I collected a corpus in which users dictated sentences and performed spoken corrections of simulated errors It was found that word or phrase size corrections were particularly hard for speech recognizers. In this talk, I will discuss efforts to improve recognition on these problematic utterances. Speaker adaptation techniques will be used to improve the recognizer’s acoustic model. Information from the correction’s context and from the user’s intial utterance will be integrated into the recognizer’s decoding network.

For further details, see our paper or see chapter 3 of my thesis

This talk is part of the Inference Group series.

Tell a friend about this talk:

This talk is included in these lists:

Note that ex-directory lists are not shown.

 

© 2006-2024 Talks.cam, University of Cambridge. Contact Us | Help and Documentation | Privacy and Publicity