Leveraging run-time feedback for efficient ASR acceleration
Document typeConference report
PublisherInstitute of Electrical and Electronics Engineers (IEEE)
Rights accessOpen Access
European Commission's projectCoCoUnit - CoCoUnit: An Energy-Efficient Processing Unit for Cognitive Computing (EC-H2020-833057)
In this work, we propose Locality-AWare-Scheme (LAWS) for an Automatic Speech Recognition (ASR) accelerator in order to significantly reduce its energy consumption and memory requirements, by leveraging the locality among consecutive segments of the speech signal. LAWS diminishes ASR's workload by up to 60% by removing most of the off-chip accesses during the ASR's decoding process. We furthermore improve LAWS's effectiveness by selectively adapting the amount of ASR's workload, based on run-time feedback. In particular, we exploit the fact that the confidence of the ASR system varies along the recognition process. When confidence is high, the ASR system can be more restrictive and reduce the amount of work. The end design provides a saving of 87% in memory requests, 2.3x reduction in energy consumption, and a speedup of 2.1x with respect to the state-of-the-art ASR accelerator.
© 2019 IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other uses, in any current or future media, including reprinting/republishing this material for advertising or promotional purposes,creating new collective works, for resale or redistribution to servers or lists, or reuse of any copyrighted component of this work in other works.
CitationYazdani, R.; Arnau, J.; Gonzalez, A. Leveraging run-time feedback for efficient ASR acceleration. A: International Conference on Parallel Architectures and Compilation. "2019 28th International Conference on Parallel Architectures and Compilation Techniques (PACT 2019): Seattle, Washington, USA: 23-26 September 2019". Institute of Electrical and Electronics Engineers (IEEE), 2019, p. 462-463.
All rights reserved. This work is protected by the corresponding intellectual and industrial property rights. Without prejudice to any existing legal exemptions, reproduction, distribution, public communication or transformation of this work are prohibited without permission of the copyright holder