Abstract | ||
---|---|---|
Reverberation leads to high word error rates (WERs) for automatic speech recognition (ASR) systems. This work presents robust acoustic features motivated by subspace modeling and human speech perception for use in large vocabulary continuous speech recognition (LVCSR). We explore different acoustic modeling strategies and language modeling techniques, and demonstrate that robust features with acoustic modeling based on deep learning can provide significant reduction in WERs in the task of recognizing reverberated speech compared to mel-cepstral features and acoustic modeling based on Gaussian Mixture Models (GMMs). |
Year | Venue | Keywords |
---|---|---|
2015 | 16TH ANNUAL CONFERENCE OF THE INTERNATIONAL SPEECH COMMUNICATION ASSOCIATION (INTERSPEECH 2015), VOLS 1-5 | deep neural networks, robust features, robust speech recognition, reverberation robustness |
Field | DocType | Citations |
Reverberation,Computer science,Speech recognition,Natural language processing,Artificial intelligence,Vocabulary | Conference | 2 |
PageRank | References | Authors |
0.39 | 0 | 7 |
Name | Order | Citations | PageRank |
---|---|---|---|
Vikramjit Mitra | 1 | 299 | 24.83 |
Julien van Hout | 2 | 54 | 6.07 |
Mitchell McLaren | 3 | 454 | 35.97 |
Wen Wang | 4 | 327 | 29.31 |
Martin Graciarena | 5 | 281 | 24.70 |
Dimitra Vergyri | 6 | 373 | 36.97 |
Horacio Franco | 7 | 543 | 72.04 |