ePrints@IIScePrints@IISc Home | About | Browse | Latest Additions | Advanced Search | Contact | Help


Kumar, R and Purushothaman, A and Sreeram, A and Ganapathy, S (2022) END-TO-END SPEECH RECOGNITION WITH JOINT DEREVERBERATION OF SUB-BAND AUTOREGRESSIVE ENVELOPES. In: 47th IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2022, 23 - 27 May 2022, Virtual, Online at Singapore, pp. 1805-1809.

IEEE_ICASSP 2022_2022_1805-1809_2022 .pdf - Published Version

Download (940kB) | Preview
Official URL: https://doi.org/10.1109/ICASSP43922.2022.9747795


The end-to-end (E2E) automatic speech recognition (ASR) systems are often required to operate in reverberant conditions, where the long-term sub-band envelopes of the speech are temporally smeared. In this paper, we develop a feature enhancement approach using a neural model operating on sub-band temporal envelopes. The temporal envelopes are modeled using the framework of frequency domain linear prediction (FDLP). The neural enhancement model proposed in this paper performs an envelope gain based enhancement of temporal envelopes. The model architecture consists of a combination of convolutional and long short term memory (LSTM) neural network layers. Further, the envelope dereverberation, feature extraction and acoustic modeling using transformer based E2E ASR can all be jointly optimized for the speech recognition task. We perform E2E speech recognition experiments on the REVERB challenge dataset as well as on the VOiCES dataset. In these experiments, the proposed joint modeling approach yields significant improvements compared to the baseline E2E ASR system (average relative improvements of 21 on the REVERB challenge dataset and about 10 on the VOiCES dataset).

Item Type: Conference Paper
Publication: ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings
Publisher: Institute of Electrical and Electronics Engineers Inc.
Additional Information: The copyright for this article belongs to the Authors.
Keywords: Acoustic Modeling; Frequency domain analysis; Long short-term memory; Multilayer neural networks; Multilayers; Network layers, Automatic speech recognition; Dereverberation; End to end; End-to-end automatic speech recognition; Frequency domain linear prediction; Frequency domains; Joint models; Linear prediction; Subbands; Temporal envelopes, Speech recognition
Department/Centre: Division of Electrical Sciences > Electrical Engineering
Date Deposited: 05 Aug 2022 09:02
Last Modified: 05 Aug 2022 09:02
URI: https://eprints.iisc.ac.in/id/eprint/75354

Actions (login required)

View Item View Item