Matches in SemOpenAlex for { <https://semopenalex.org/work/W4233670287> ?p ?o ?g. }
Showing items 1 to 97 of
97
with 100 items per page.
- W4233670287 abstract "Automatic Speech Recognition (ASR) is becoming increasingly ubiquitous, especially in the mobile segment. Fast and accurate ASR comes at a high energy cost which is not affordable for the tiny power budget of mobile devices. Hardware acceleration can reduce power consumption of ASR systems, while delivering high-performance. In this paper, we present an accelerator for large-vocabulary, speaker-independent, continuous speech recognition. It focuses on the Viterbi search algorithm, that represents the main bottleneck in an ASR system. The proposed design includes innovative techniques to improve the memory subsystem, since memory is identified as the main bottleneck for performance and power in the design of these accelerators. We propose a prefetching scheme tailored to the needs of an ASR system that hides main memory latency for a large fraction of the memory accesses with a negligible impact on area. In addition, we introduce a novel bandwidth saving technique that removes 20% of the off-chip memory accesses issued during the Viterbi search. The proposed design outperforms software implementations running on the CPU by orders of magnitude and achieves 1.7x speedup over a highly optimized CUDA implementation running on a high-end Geforce GTX 980 GPU, while reducing by two orders of magnitude (287x) the energy required to convert the speech into text." @default.
- W4233670287 created "2022-05-12" @default.
- W4233670287 creator A5032189355 @default.
- W4233670287 creator A5047859397 @default.
- W4233670287 creator A5062132797 @default.
- W4233670287 creator A5065422455 @default.
- W4233670287 date "2016-10-01" @default.
- W4233670287 modified "2023-10-09" @default.
- W4233670287 title "An ultra low-power hardware accelerator for automatic speech recognition" @default.
- W4233670287 cites W1485823550 @default.
- W4233670287 cites W1494198834 @default.
- W4233670287 cites W1802482518 @default.
- W4233670287 cites W1969648707 @default.
- W4233670287 cites W1986989509 @default.
- W4233670287 cites W2023619619 @default.
- W4233670287 cites W2034227610 @default.
- W4233670287 cites W2041414040 @default.
- W4233670287 cites W2046526098 @default.
- W4233670287 cites W2046932483 @default.
- W4233670287 cites W2047417997 @default.
- W4233670287 cites W2048266589 @default.
- W4233670287 cites W2049553000 @default.
- W4233670287 cites W2067523571 @default.
- W4233670287 cites W2101328377 @default.
- W4233670287 cites W2141493634 @default.
- W4233670287 cites W2142801765 @default.
- W4233670287 cites W2160688560 @default.
- W4233670287 cites W2170382128 @default.
- W4233670287 cites W2339460208 @default.
- W4233670287 cites W2537450429 @default.
- W4233670287 cites W4255238016 @default.
- W4233670287 cites W954411423 @default.
- W4233670287 doi "https://doi.org/10.1109/micro.2016.7783750" @default.
- W4233670287 hasPublicationYear "2016" @default.
- W4233670287 type Work @default.
- W4233670287 citedByCount "19" @default.
- W4233670287 countsByYear W42336702872017 @default.
- W4233670287 countsByYear W42336702872018 @default.
- W4233670287 countsByYear W42336702872019 @default.
- W4233670287 countsByYear W42336702872020 @default.
- W4233670287 countsByYear W42336702872021 @default.
- W4233670287 countsByYear W42336702872022 @default.
- W4233670287 crossrefType "proceedings-article" @default.
- W4233670287 hasAuthorship W4233670287A5032189355 @default.
- W4233670287 hasAuthorship W4233670287A5047859397 @default.
- W4233670287 hasAuthorship W4233670287A5062132797 @default.
- W4233670287 hasAuthorship W4233670287A5065422455 @default.
- W4233670287 hasBestOaLocation W42336702872 @default.
- W4233670287 hasConcept C13164978 @default.
- W4233670287 hasConcept C149635348 @default.
- W4233670287 hasConcept C173608175 @default.
- W4233670287 hasConcept C18903297 @default.
- W4233670287 hasConcept C23224414 @default.
- W4233670287 hasConcept C2780165032 @default.
- W4233670287 hasConcept C2780513914 @default.
- W4233670287 hasConcept C2781357197 @default.
- W4233670287 hasConcept C28490314 @default.
- W4233670287 hasConcept C41008148 @default.
- W4233670287 hasConcept C42935608 @default.
- W4233670287 hasConcept C60582962 @default.
- W4233670287 hasConcept C68339613 @default.
- W4233670287 hasConcept C82687282 @default.
- W4233670287 hasConcept C86803240 @default.
- W4233670287 hasConcept C9390403 @default.
- W4233670287 hasConceptScore W4233670287C13164978 @default.
- W4233670287 hasConceptScore W4233670287C149635348 @default.
- W4233670287 hasConceptScore W4233670287C173608175 @default.
- W4233670287 hasConceptScore W4233670287C18903297 @default.
- W4233670287 hasConceptScore W4233670287C23224414 @default.
- W4233670287 hasConceptScore W4233670287C2780165032 @default.
- W4233670287 hasConceptScore W4233670287C2780513914 @default.
- W4233670287 hasConceptScore W4233670287C2781357197 @default.
- W4233670287 hasConceptScore W4233670287C28490314 @default.
- W4233670287 hasConceptScore W4233670287C41008148 @default.
- W4233670287 hasConceptScore W4233670287C42935608 @default.
- W4233670287 hasConceptScore W4233670287C60582962 @default.
- W4233670287 hasConceptScore W4233670287C68339613 @default.
- W4233670287 hasConceptScore W4233670287C82687282 @default.
- W4233670287 hasConceptScore W4233670287C86803240 @default.
- W4233670287 hasConceptScore W4233670287C9390403 @default.
- W4233670287 hasLocation W42336702871 @default.
- W4233670287 hasLocation W42336702872 @default.
- W4233670287 hasOpenAccess W4233670287 @default.
- W4233670287 hasPrimaryLocation W42336702871 @default.
- W4233670287 hasRelatedWork W1508811950 @default.
- W4233670287 hasRelatedWork W2074831258 @default.
- W4233670287 hasRelatedWork W2373066471 @default.
- W4233670287 hasRelatedWork W3094426418 @default.
- W4233670287 hasRelatedWork W3146496550 @default.
- W4233670287 hasRelatedWork W4241632897 @default.
- W4233670287 hasRelatedWork W4312653389 @default.
- W4233670287 hasRelatedWork W4312655603 @default.
- W4233670287 hasRelatedWork W4366380722 @default.
- W4233670287 hasRelatedWork W4377969791 @default.
- W4233670287 isParatext "false" @default.
- W4233670287 isRetracted "false" @default.
- W4233670287 workType "article" @default.