Matches in SemOpenAlex for { <https://semopenalex.org/work/W4319862416> ?p ?o ?g. }
Showing items 1 to 83 of
83
with 100 items per page.
- W4319862416 abstract "While Self-Supervised Learning has helped reap the benefit of the scale from the available unlabeled data, the learning paradigms are continously being bettered. We present a new pre-training strategy named ccc-wav2vec 2.0, which uses clustering and an augmentation based cross-contrastive loss as its self-supervised objective. Through the clustering module we scale down the influence of those negative examples that are highly similar to the positive. The Cross-Contrastive loss is computed between the encoder output of the original sample and the quantizer output of its augmentation, and vice-versa, bringing robustness to the pre-training strategy. ccc-wav2vec 2.0 achieves upto 15.6% and 12.7% relative WER improvement over the baseline wav2vec 2.0 on the test-clean and test-other sets respectively of LibriSpeech, without the use of any language model. The proposed method also achieves upto 14.9% relative WER improvement over the baseline wav2vec 2.0, when fine-tuned on Switchboard data." @default.
- W4319862416 created "2023-02-11" @default.
- W4319862416 creator A5033408639 @default.
- W4319862416 creator A5075877984 @default.
- W4319862416 creator A5085660348 @default.
- W4319862416 date "2023-01-09" @default.
- W4319862416 modified "2023-09-26" @default.
- W4319862416 title "CCC-WAV2VEC 2.0: Clustering AIDED Cross Contrastive Self-Supervised Learning of Speech Representations" @default.
- W4319862416 cites W1494198834 @default.
- W4319862416 cites W2696967604 @default.
- W4319862416 cites W2933138175 @default.
- W4319862416 cites W2936774411 @default.
- W4319862416 cites W2948638722 @default.
- W4319862416 cites W2982223350 @default.
- W4319862416 cites W3024869864 @default.
- W4319862416 cites W3041561163 @default.
- W4319862416 cites W3087124270 @default.
- W4319862416 cites W3144810982 @default.
- W4319862416 cites W3154677174 @default.
- W4319862416 cites W3163839574 @default.
- W4319862416 cites W3197580070 @default.
- W4319862416 cites W3206252155 @default.
- W4319862416 cites W3209059054 @default.
- W4319862416 cites W3209984917 @default.
- W4319862416 cites W4221163898 @default.
- W4319862416 cites W4226033575 @default.
- W4319862416 cites W4297841844 @default.
- W4319862416 doi "https://doi.org/10.1109/slt54892.2023.10022552" @default.
- W4319862416 hasPublicationYear "2023" @default.
- W4319862416 type Work @default.
- W4319862416 citedByCount "0" @default.
- W4319862416 crossrefType "proceedings-article" @default.
- W4319862416 hasAuthorship W4319862416A5033408639 @default.
- W4319862416 hasAuthorship W4319862416A5075877984 @default.
- W4319862416 hasAuthorship W4319862416A5085660348 @default.
- W4319862416 hasConcept C104317684 @default.
- W4319862416 hasConcept C111919701 @default.
- W4319862416 hasConcept C118505674 @default.
- W4319862416 hasConcept C119857082 @default.
- W4319862416 hasConcept C153180895 @default.
- W4319862416 hasConcept C154945302 @default.
- W4319862416 hasConcept C16910744 @default.
- W4319862416 hasConcept C185592680 @default.
- W4319862416 hasConcept C199360897 @default.
- W4319862416 hasConcept C204321447 @default.
- W4319862416 hasConcept C28490314 @default.
- W4319862416 hasConcept C41008148 @default.
- W4319862416 hasConcept C51632099 @default.
- W4319862416 hasConcept C55493867 @default.
- W4319862416 hasConcept C63479239 @default.
- W4319862416 hasConcept C73555534 @default.
- W4319862416 hasConceptScore W4319862416C104317684 @default.
- W4319862416 hasConceptScore W4319862416C111919701 @default.
- W4319862416 hasConceptScore W4319862416C118505674 @default.
- W4319862416 hasConceptScore W4319862416C119857082 @default.
- W4319862416 hasConceptScore W4319862416C153180895 @default.
- W4319862416 hasConceptScore W4319862416C154945302 @default.
- W4319862416 hasConceptScore W4319862416C16910744 @default.
- W4319862416 hasConceptScore W4319862416C185592680 @default.
- W4319862416 hasConceptScore W4319862416C199360897 @default.
- W4319862416 hasConceptScore W4319862416C204321447 @default.
- W4319862416 hasConceptScore W4319862416C28490314 @default.
- W4319862416 hasConceptScore W4319862416C41008148 @default.
- W4319862416 hasConceptScore W4319862416C51632099 @default.
- W4319862416 hasConceptScore W4319862416C55493867 @default.
- W4319862416 hasConceptScore W4319862416C63479239 @default.
- W4319862416 hasConceptScore W4319862416C73555534 @default.
- W4319862416 hasLocation W43198624161 @default.
- W4319862416 hasOpenAccess W4319862416 @default.
- W4319862416 hasPrimaryLocation W43198624161 @default.
- W4319862416 hasRelatedWork W105527591 @default.
- W4319862416 hasRelatedWork W2945374320 @default.
- W4319862416 hasRelatedWork W2949412037 @default.
- W4319862416 hasRelatedWork W2996325578 @default.
- W4319862416 hasRelatedWork W3035729345 @default.
- W4319862416 hasRelatedWork W3094805729 @default.
- W4319862416 hasRelatedWork W3194633786 @default.
- W4319862416 hasRelatedWork W4288350321 @default.
- W4319862416 hasRelatedWork W4312741812 @default.
- W4319862416 hasRelatedWork W4320472933 @default.
- W4319862416 isParatext "false" @default.
- W4319862416 isRetracted "false" @default.
- W4319862416 workType "article" @default.