Paper Info
Title
REPRESENTATION LEARNING FOR SPEECH RECOGNITION USING FEEDBACK BASED RELEVANCE WEIGHTING
Abstract
In this work, we propose an acoustic embedding based approach for representation learning in speech recognition. The proposed approach involves two stages comprising of acoustic filterbank learning from raw waveform, followed by modulation filterbank learning. In each stage, a relevance weighting operation is employed that acts as a feature selection module. In particular, the relevance weighting network receives embeddings of the model outputs from the previous time instants as feedback. The proposed relevance weighting scheme allows the respective feature representations to be adaptively selected before propagation to the higher layers. The application of the proposed approach for the task of speech recognition on Aurora-4 and CHiME-3 datasets gives significant performance improvements over baseline systems on raw waveform signal as well as those based on mel representations (average relative improvement of 15% over the mel baseline on Aurora-4 dataset and 7% on CHiME-3 dataset).
Year
DOI
Venue
2021
10.1109/ICASSP39728.2021.9414649
2021 IEEE INTERNATIONAL CONFERENCE ON ACOUSTICS, SPEECH AND SIGNAL PROCESSING (ICASSP 2021)
Keywords
DocType
ISSN
Speech representation learning, feedback of acoustic embeddings, raw speech waveform, 2-stage relevance weighting, speech recognition
Conference
IEEE International Conference on Acoustics, Speech, & Signal Processing (ICASSP) 2021
Citations 
PageRank 
References 
0
0.34
0
Authors
2
Name
Order
Citations
PageRank
Purvi Agrawal122.38
Sriram Ganapathy225239.62