Some new approaches to speech recognition and language modelling
Add to your list(s)
Download to your calendar using vCal
If you have a question about this talk, please contact Kai Yu.
In the first part of the talk, I will review some of the research at the University of Toronto on the use of deep belief nets for speech recognition. Then I will describe some very recent work that uses a rather different approach to learn acoustic events that have precise times and amplitudes. Finally I will briefly describe a much better way of optimizing recurrent neural networks. When applied to a novel type of recurrent neural net, this new optimizer learns extremely good models of character sequences. This is joint work with George Dahl, Abdel-rahman Mohamed, Navdeep Jaitly, James Martens and Ilya Sutskever.
This talk is part of the Machine Intelligence Lab Seminar series.
This talk is included in these lists:
Note that ex-directory lists are not shown.
|