Matches in SemOpenAlex for { <https://semopenalex.org/work/W3131305672> ?p ?o ?g. }
Showing items 1 to 95 of
95
with 100 items per page.
- W3131305672 abstract "End-to-end models for robust automatic speech recognition (ASR) have not been sufficiently well-explored in prior work. With end-to-end models, one could choose to preprocess the input speech using speech enhancement techniques and train the model using enhanced speech. Another alternative is to pass the noisy speech as input and modify the model architecture to adapt to noisy speech. A systematic comparison of these two approaches for end-to-end robust ASR has not been attempted before. We address this gap and present a detailed comparison of speech enhancement-based techniques and three different model-based adaptation techniques covering data augmentation, multi-task learning, and adversarial learning for robust ASR. While adversarial learning is the best-performing technique on certain noise types, it comes at the cost of degrading clean speech WER. On other relatively stationary noise types, a new speech enhancement technique outperformed all the model-based adaptation techniques. This suggests that knowledge of the underlying noise type can meaningfully inform the choice of adaptation technique." @default.
- W3131305672 created "2021-03-01" @default.
- W3131305672 creator A5036738038 @default.
- W3131305672 creator A5039920828 @default.
- W3131305672 creator A5060996961 @default.
- W3131305672 date "2021-02-11" @default.
- W3131305672 modified "2023-09-27" @default.
- W3131305672 title "An Investigation of End-to-End Models for Robust Speech Recognition" @default.
- W3131305672 cites W1494198834 @default.
- W3131305672 cites W2033875152 @default.
- W3131305672 cites W2042141988 @default.
- W3131305672 cites W2062164080 @default.
- W3131305672 cites W2193413348 @default.
- W3131305672 cites W2219249508 @default.
- W3131305672 cites W2587717635 @default.
- W3131305672 cites W2618099328 @default.
- W3131305672 cites W2696967604 @default.
- W3131305672 cites W2885724687 @default.
- W3131305672 cites W2950333269 @default.
- W3131305672 cites W2963341071 @default.
- W3131305672 cites W2963364041 @default.
- W3131305672 cites W2963826681 @default.
- W3131305672 cites W2972320711 @default.
- W3131305672 cites W2973188947 @default.
- W3131305672 cites W3015411292 @default.
- W3131305672 cites W3034273309 @default.
- W3131305672 cites W3097945073 @default.
- W3131305672 cites W3209141406 @default.
- W3131305672 hasPublicationYear "2021" @default.
- W3131305672 type Work @default.
- W3131305672 sameAs 3131305672 @default.
- W3131305672 citedByCount "0" @default.
- W3131305672 crossrefType "posted-content" @default.
- W3131305672 hasAuthorship W3131305672A5036738038 @default.
- W3131305672 hasAuthorship W3131305672A5039920828 @default.
- W3131305672 hasAuthorship W3131305672A5060996961 @default.
- W3131305672 hasConcept C115961682 @default.
- W3131305672 hasConcept C120665830 @default.
- W3131305672 hasConcept C121332964 @default.
- W3131305672 hasConcept C127413603 @default.
- W3131305672 hasConcept C139807058 @default.
- W3131305672 hasConcept C154945302 @default.
- W3131305672 hasConcept C163294075 @default.
- W3131305672 hasConcept C201995342 @default.
- W3131305672 hasConcept C204201278 @default.
- W3131305672 hasConcept C2776182073 @default.
- W3131305672 hasConcept C2780451532 @default.
- W3131305672 hasConcept C28490314 @default.
- W3131305672 hasConcept C41008148 @default.
- W3131305672 hasConcept C61328038 @default.
- W3131305672 hasConcept C74296488 @default.
- W3131305672 hasConcept C99498987 @default.
- W3131305672 hasConceptScore W3131305672C115961682 @default.
- W3131305672 hasConceptScore W3131305672C120665830 @default.
- W3131305672 hasConceptScore W3131305672C121332964 @default.
- W3131305672 hasConceptScore W3131305672C127413603 @default.
- W3131305672 hasConceptScore W3131305672C139807058 @default.
- W3131305672 hasConceptScore W3131305672C154945302 @default.
- W3131305672 hasConceptScore W3131305672C163294075 @default.
- W3131305672 hasConceptScore W3131305672C201995342 @default.
- W3131305672 hasConceptScore W3131305672C204201278 @default.
- W3131305672 hasConceptScore W3131305672C2776182073 @default.
- W3131305672 hasConceptScore W3131305672C2780451532 @default.
- W3131305672 hasConceptScore W3131305672C28490314 @default.
- W3131305672 hasConceptScore W3131305672C41008148 @default.
- W3131305672 hasConceptScore W3131305672C61328038 @default.
- W3131305672 hasConceptScore W3131305672C74296488 @default.
- W3131305672 hasConceptScore W3131305672C99498987 @default.
- W3131305672 hasLocation W31313056721 @default.
- W3131305672 hasOpenAccess W3131305672 @default.
- W3131305672 hasPrimaryLocation W31313056721 @default.
- W3131305672 hasRelatedWork W141701120 @default.
- W3131305672 hasRelatedWork W1504449689 @default.
- W3131305672 hasRelatedWork W2124338691 @default.
- W3131305672 hasRelatedWork W2152305730 @default.
- W3131305672 hasRelatedWork W2572023240 @default.
- W3131305672 hasRelatedWork W2892204792 @default.
- W3131305672 hasRelatedWork W2935938411 @default.
- W3131305672 hasRelatedWork W2963321191 @default.
- W3131305672 hasRelatedWork W2963954250 @default.
- W3131305672 hasRelatedWork W2964182350 @default.
- W3131305672 hasRelatedWork W2972592847 @default.
- W3131305672 hasRelatedWork W3011500442 @default.
- W3131305672 hasRelatedWork W3031135612 @default.
- W3131305672 hasRelatedWork W3094225009 @default.
- W3131305672 hasRelatedWork W3116834994 @default.
- W3131305672 hasRelatedWork W3163142165 @default.
- W3131305672 hasRelatedWork W3180992661 @default.
- W3131305672 hasRelatedWork W3186546663 @default.
- W3131305672 hasRelatedWork W3199713199 @default.
- W3131305672 hasRelatedWork W3209376089 @default.
- W3131305672 isParatext "false" @default.
- W3131305672 isRetracted "false" @default.
- W3131305672 magId "3131305672" @default.
- W3131305672 workType "article" @default.