Low-Rank Representation of Nearest Neighbor Phone Posterior Probabilities to Enhance DNN Acoustic Modeling

We hypothesize that optimal deep neural networks (DNN) class-conditional posterior probabilities live in a union of low-dimensional subspaces. In real test conditions, DNN posteriors encode uncertainties which can be regarded as a superposition of unstructured sparse noise to the optimal posteriors. We aim to investigate different ways to structure the DNN outputs exploiting low-rank representation (LRR) techniques. Using a large number of training posterior vectors, the underlying low-dimensional subspace is identified through nearest neighbor analysis, and low-rank decomposition enables separation of the ``optimal'' posteriors from the spurious uncertainties at the DNN output. Experiments demonstrate that by processing subsets of posteriors which possess strong subspace similarity, low-rank representation enables enhancement of posterior probabilities, and lead to higher speech recognition accuracy based on the hybrid DNN-hidden Markov model (HMM) system.


Year:
2016
Publisher:
Idiap
Keywords:
Laboratories:




 Record created 2016-04-19, last modified 2018-03-17

n/a:
LRR - Download fulltextPDF
Luyet_Idiap-RR-04-2016 - Download fulltextPDF
Rate this document:

Rate this document:
1
2
3
 
(Not yet reviewed)