pith. sign in

arxiv: 1803.09816 · v1 · pith:A4OMGAELnew · submitted 2018-03-26 · 💻 cs.SD · cs.CL· eess.AS

Spectral feature mapping with mimic loss for robust speech recognition

classification 💻 cs.SD cs.CLeess.AS
keywords speechspectralclassifierde-noisedcleancriterionenhancerlocal
0
0 comments X
read the original abstract

For the task of speech enhancement, local learning objectives are agnostic to phonetic structures helpful for speech recognition. We propose to add a global criterion to ensure de-noised speech is useful for downstream tasks like ASR. We first train a spectral classifier on clean speech to predict senone labels. Then, the spectral classifier is joined with our speech enhancer as a noisy speech recognizer. This model is taught to imitate the output of the spectral classifier alone on clean speech. This \textit{mimic loss} is combined with the traditional local criterion to train the speech enhancer to produce de-noised speech. Feeding the de-noised speech to an off-the-shelf Kaldi training recipe for the CHiME-2 corpus shows significant improvements in WER.

This paper has not been read by Pith yet.

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.