Matches in SemOpenAlex for { <https://semopenalex.org/work/W3146125295> ?p ?o ?g. }
Showing items 1 to 85 of
85
with 100 items per page.
- W3146125295 abstract "Whispering is an important mode of human speech, but no end-to-end recognition results for it were reported yet, probably due to the scarcity of available whispered speech data. In this paper, we present several approaches for end-to-end (E2E) recognition of whispered speech considering the special characteristics of whispered speech and the scarcity of data. This includes a frequency-weighted SpecAugment policy and a frequency-divided CNN feature extractor for better capturing the high-frequency structures of whispered speech, and a layer-wise transfer learning approach to pre-train a model with normal or normal-to-whispered converted speech then fine-tune it with whispered speech to bridge the gap between whispered and normal speech. We achieve an overall relative reduction of 19.8% in PER and 44.4% in CER on a relatively small whispered TIMIT corpus. The results indicate as long as we have a good E2E model pre-trained on normal or pseudo-whispered speech, a relatively small set of whispered speech may suffice to obtain a reasonably good E2E whispered speech recognizer." @default.
- W3146125295 created "2021-04-13" @default.
- W3146125295 creator A5039767083 @default.
- W3146125295 creator A5040508737 @default.
- W3146125295 creator A5044010123 @default.
- W3146125295 creator A5078976109 @default.
- W3146125295 date "2021-01-19" @default.
- W3146125295 modified "2023-09-26" @default.
- W3146125295 title "End-to-End Whispered Speech Recognition with Frequency-Weighted Approaches and Pseudo Whisper Pre-training" @default.
- W3146125295 cites W1494198834 @default.
- W3146125295 cites W1985258458 @default.
- W3146125295 cites W2022047569 @default.
- W3146125295 cites W2041470347 @default.
- W3146125295 cites W2119648481 @default.
- W3146125295 cites W2125838338 @default.
- W3146125295 cites W2127141656 @default.
- W3146125295 cites W2144994235 @default.
- W3146125295 cites W2146618065 @default.
- W3146125295 cites W2160815625 @default.
- W3146125295 cites W2303272654 @default.
- W3146125295 cites W2327501763 @default.
- W3146125295 cites W242941288 @default.
- W3146125295 cites W2437658533 @default.
- W3146125295 cites W2516966418 @default.
- W3146125295 cites W2530876040 @default.
- W3146125295 cites W2608506766 @default.
- W3146125295 cites W2627092829 @default.
- W3146125295 cites W2767064007 @default.
- W3146125295 cites W2935803161 @default.
- W3146125295 cites W2936774411 @default.
- W3146125295 cites W2940180244 @default.
- W3146125295 cites W2962824709 @default.
- W3146125295 cites W2963303951 @default.
- W3146125295 cites W2972924908 @default.
- W3146125295 cites W2972970915 @default.
- W3146125295 cites W3006827623 @default.
- W3146125295 cites W3015889230 @default.
- W3146125295 doi "https://doi.org/10.1109/slt48900.2021.9383595" @default.
- W3146125295 hasPublicationYear "2021" @default.
- W3146125295 type Work @default.
- W3146125295 sameAs 3146125295 @default.
- W3146125295 citedByCount "1" @default.
- W3146125295 countsByYear W31461252952022 @default.
- W3146125295 crossrefType "proceedings-article" @default.
- W3146125295 hasAuthorship W3146125295A5039767083 @default.
- W3146125295 hasAuthorship W3146125295A5040508737 @default.
- W3146125295 hasAuthorship W3146125295A5044010123 @default.
- W3146125295 hasAuthorship W3146125295A5078976109 @default.
- W3146125295 hasBestOaLocation W31461252952 @default.
- W3146125295 hasConcept C138885662 @default.
- W3146125295 hasConcept C204201278 @default.
- W3146125295 hasConcept C23224414 @default.
- W3146125295 hasConcept C2776401178 @default.
- W3146125295 hasConcept C2778724510 @default.
- W3146125295 hasConcept C28490314 @default.
- W3146125295 hasConcept C41008148 @default.
- W3146125295 hasConcept C41895202 @default.
- W3146125295 hasConcept C61328038 @default.
- W3146125295 hasConceptScore W3146125295C138885662 @default.
- W3146125295 hasConceptScore W3146125295C204201278 @default.
- W3146125295 hasConceptScore W3146125295C23224414 @default.
- W3146125295 hasConceptScore W3146125295C2776401178 @default.
- W3146125295 hasConceptScore W3146125295C2778724510 @default.
- W3146125295 hasConceptScore W3146125295C28490314 @default.
- W3146125295 hasConceptScore W3146125295C41008148 @default.
- W3146125295 hasConceptScore W3146125295C41895202 @default.
- W3146125295 hasConceptScore W3146125295C61328038 @default.
- W3146125295 hasLocation W31461252951 @default.
- W3146125295 hasLocation W31461252952 @default.
- W3146125295 hasOpenAccess W3146125295 @default.
- W3146125295 hasPrimaryLocation W31461252951 @default.
- W3146125295 hasRelatedWork W170831052 @default.
- W3146125295 hasRelatedWork W1980604799 @default.
- W3146125295 hasRelatedWork W2009810445 @default.
- W3146125295 hasRelatedWork W2100854157 @default.
- W3146125295 hasRelatedWork W2107338293 @default.
- W3146125295 hasRelatedWork W2124093511 @default.
- W3146125295 hasRelatedWork W2390182490 @default.
- W3146125295 hasRelatedWork W2978471304 @default.
- W3146125295 hasRelatedWork W4221032780 @default.
- W3146125295 hasRelatedWork W4247725880 @default.
- W3146125295 isParatext "false" @default.
- W3146125295 isRetracted "false" @default.
- W3146125295 magId "3146125295" @default.
- W3146125295 workType "article" @default.