Matches in SemOpenAlex for { <https://semopenalex.org/work/W4312120634> ?p ?o ?g. }
Showing items 1 to 100 of
100
with 100 items per page.
- W4312120634 abstract "In the field of deep learning, researchers often focus on inventing novel neural network models and improving benchmarks. In contrast, application developers are interested in making models suitable for actual products, which involves optimizing a model for faster inference and adapting a model to various platforms (e.g., C++ and Python). In this work, to fill the gap between the two, we establish an effective procedure for optimizing a PyTorch-based research-oriented model for deployment, taking ESPnet, a widely used toolkit for speech processing, as an instance. We introduce different techniques to ESPnet, including converting a model into an ONNX format, fusing nodes in a graph, and quantizing parameters, which lead to approximately 1.3-2x speedup in various tasks (i.e., ASR, TTS, speech translation, and spoken language understanding) while keeping its performance without any additional training. Our ESPnet-ONNX will be publicly available at https://github.com/espnet/espnet_onnx." @default.
- W4312120634 created "2023-01-04" @default.
- W4312120634 creator A5001291873 @default.
- W4312120634 creator A5030925462 @default.
- W4312120634 creator A5078778981 @default.
- W4312120634 creator A5090323167 @default.
- W4312120634 date "2022-11-07" @default.
- W4312120634 modified "2023-09-27" @default.
- W4312120634 title "ESPnet-ONNX: Bridging a Gap Between Research and Production" @default.
- W4312120634 cites W2043701535 @default.
- W4312120634 cites W2058094241 @default.
- W4312120634 cites W2889129739 @default.
- W4312120634 cites W2892008152 @default.
- W4312120634 cites W2946794439 @default.
- W4312120634 cites W2962780374 @default.
- W4312120634 cites W2963925437 @default.
- W4312120634 cites W2964110616 @default.
- W4312120634 cites W2964243274 @default.
- W4312120634 cites W2972778435 @default.
- W4312120634 cites W3037217258 @default.
- W4312120634 cites W3095552229 @default.
- W4312120634 cites W3097777922 @default.
- W4312120634 cites W3097882114 @default.
- W4312120634 cites W3100460087 @default.
- W4312120634 cites W3112157188 @default.
- W4312120634 cites W3160799772 @default.
- W4312120634 cites W3162919436 @default.
- W4312120634 cites W3163793923 @default.
- W4312120634 cites W3197148831 @default.
- W4312120634 cites W3197478142 @default.
- W4312120634 cites W3197580070 @default.
- W4312120634 cites W3206573929 @default.
- W4312120634 cites W3217016993 @default.
- W4312120634 cites W3217767527 @default.
- W4312120634 cites W4221138270 @default.
- W4312120634 cites W4224612669 @default.
- W4312120634 cites W4224918069 @default.
- W4312120634 cites W4285188582 @default.
- W4312120634 cites W4297841482 @default.
- W4312120634 cites W4312825166 @default.
- W4312120634 doi "https://doi.org/10.23919/apsipaasc55919.2022.9979824" @default.
- W4312120634 hasPublicationYear "2022" @default.
- W4312120634 type Work @default.
- W4312120634 citedByCount "0" @default.
- W4312120634 crossrefType "proceedings-article" @default.
- W4312120634 hasAuthorship W4312120634A5001291873 @default.
- W4312120634 hasAuthorship W4312120634A5030925462 @default.
- W4312120634 hasAuthorship W4312120634A5078778981 @default.
- W4312120634 hasAuthorship W4312120634A5090323167 @default.
- W4312120634 hasBestOaLocation W43121206342 @default.
- W4312120634 hasConcept C105339364 @default.
- W4312120634 hasConcept C108583219 @default.
- W4312120634 hasConcept C115903868 @default.
- W4312120634 hasConcept C154945302 @default.
- W4312120634 hasConcept C173608175 @default.
- W4312120634 hasConcept C174348530 @default.
- W4312120634 hasConcept C199360897 @default.
- W4312120634 hasConcept C203005215 @default.
- W4312120634 hasConcept C204321447 @default.
- W4312120634 hasConcept C2776214188 @default.
- W4312120634 hasConcept C2986862884 @default.
- W4312120634 hasConcept C31258907 @default.
- W4312120634 hasConcept C41008148 @default.
- W4312120634 hasConcept C50644808 @default.
- W4312120634 hasConcept C519991488 @default.
- W4312120634 hasConcept C68339613 @default.
- W4312120634 hasConceptScore W4312120634C105339364 @default.
- W4312120634 hasConceptScore W4312120634C108583219 @default.
- W4312120634 hasConceptScore W4312120634C115903868 @default.
- W4312120634 hasConceptScore W4312120634C154945302 @default.
- W4312120634 hasConceptScore W4312120634C173608175 @default.
- W4312120634 hasConceptScore W4312120634C174348530 @default.
- W4312120634 hasConceptScore W4312120634C199360897 @default.
- W4312120634 hasConceptScore W4312120634C203005215 @default.
- W4312120634 hasConceptScore W4312120634C204321447 @default.
- W4312120634 hasConceptScore W4312120634C2776214188 @default.
- W4312120634 hasConceptScore W4312120634C2986862884 @default.
- W4312120634 hasConceptScore W4312120634C31258907 @default.
- W4312120634 hasConceptScore W4312120634C41008148 @default.
- W4312120634 hasConceptScore W4312120634C50644808 @default.
- W4312120634 hasConceptScore W4312120634C519991488 @default.
- W4312120634 hasConceptScore W4312120634C68339613 @default.
- W4312120634 hasFunder F4320306076 @default.
- W4312120634 hasLocation W43121206341 @default.
- W4312120634 hasLocation W43121206342 @default.
- W4312120634 hasOpenAccess W4312120634 @default.
- W4312120634 hasPrimaryLocation W43121206341 @default.
- W4312120634 hasRelatedWork W1539505509 @default.
- W4312120634 hasRelatedWork W1585034923 @default.
- W4312120634 hasRelatedWork W1985007624 @default.
- W4312120634 hasRelatedWork W2327204559 @default.
- W4312120634 hasRelatedWork W2529681551 @default.
- W4312120634 hasRelatedWork W2989698750 @default.
- W4312120634 hasRelatedWork W3092111461 @default.
- W4312120634 hasRelatedWork W3107474891 @default.
- W4312120634 hasRelatedWork W3198474835 @default.
- W4312120634 hasRelatedWork W4287646082 @default.
- W4312120634 isParatext "false" @default.
- W4312120634 isRetracted "false" @default.
- W4312120634 workType "article" @default.