Matches in SemOpenAlex for { <https://semopenalex.org/work/W4385570104> ?p ?o ?g. }
Showing items 1 to 49 of
49
with 100 items per page.
- W4385570104 abstract "Automatic Speech Recognition (ASR) is essential for any voice-based application. The streaming capability of ASR becomes necessary to provide immediate feedback to the user in applications like Voice Search. LSTM/RNN and CTC based ASR systems are very simple to train and deploy for low latency streaming applications but have lower accuracy when compared to the state-of-the-art models. In this work, we build accurate LSTM, attention and CTC based streaming ASR models for large-scale Hinglish (blend of Hindi and English) Voice Search. We evaluate how various modifications in vanilla LSTM training improve the system’s accuracy while preserving the streaming capabilities. We also discuss a simple integration of end-of-speech (EOS) detection with CTC models, which helps reduce the overall search latency. Our model achieves a word error rate (WER) of 3.69% without EOS and 4.78% with EOS, with ~1300 ms (~46.64%) reduction in latency." @default.
- W4385570104 created "2023-08-05" @default.
- W4385570104 creator A5052311168 @default.
- W4385570104 creator A5080735926 @default.
- W4385570104 date "2023-01-01" @default.
- W4385570104 modified "2023-09-24" @default.
- W4385570104 title "Building Accurate Low Latency ASR for Streaming Voice Search in E-commerce" @default.
- W4385570104 doi "https://doi.org/10.18653/v1/2023.acl-industry.26" @default.
- W4385570104 hasPublicationYear "2023" @default.
- W4385570104 type Work @default.
- W4385570104 citedByCount "0" @default.
- W4385570104 crossrefType "proceedings-article" @default.
- W4385570104 hasAuthorship W4385570104A5052311168 @default.
- W4385570104 hasAuthorship W4385570104A5080735926 @default.
- W4385570104 hasBestOaLocation W43855701041 @default.
- W4385570104 hasConcept C154945302 @default.
- W4385570104 hasConcept C28490314 @default.
- W4385570104 hasConcept C31258907 @default.
- W4385570104 hasConcept C40969351 @default.
- W4385570104 hasConcept C41008148 @default.
- W4385570104 hasConcept C46637626 @default.
- W4385570104 hasConcept C519982507 @default.
- W4385570104 hasConcept C76155785 @default.
- W4385570104 hasConcept C82876162 @default.
- W4385570104 hasConceptScore W4385570104C154945302 @default.
- W4385570104 hasConceptScore W4385570104C28490314 @default.
- W4385570104 hasConceptScore W4385570104C31258907 @default.
- W4385570104 hasConceptScore W4385570104C40969351 @default.
- W4385570104 hasConceptScore W4385570104C41008148 @default.
- W4385570104 hasConceptScore W4385570104C46637626 @default.
- W4385570104 hasConceptScore W4385570104C519982507 @default.
- W4385570104 hasConceptScore W4385570104C76155785 @default.
- W4385570104 hasConceptScore W4385570104C82876162 @default.
- W4385570104 hasLocation W43855701041 @default.
- W4385570104 hasOpenAccess W4385570104 @default.
- W4385570104 hasPrimaryLocation W43855701041 @default.
- W4385570104 hasRelatedWork W1574680451 @default.
- W4385570104 hasRelatedWork W1982907196 @default.
- W4385570104 hasRelatedWork W2084242067 @default.
- W4385570104 hasRelatedWork W2108827983 @default.
- W4385570104 hasRelatedWork W2226826524 @default.
- W4385570104 hasRelatedWork W2308409568 @default.
- W4385570104 hasRelatedWork W2789982723 @default.
- W4385570104 hasRelatedWork W2909132900 @default.
- W4385570104 hasRelatedWork W3199218023 @default.
- W4385570104 hasRelatedWork W4309224965 @default.
- W4385570104 isParatext "false" @default.
- W4385570104 isRetracted "false" @default.
- W4385570104 workType "article" @default.