Matches in SemOpenAlex for { <https://semopenalex.org/work/W4368754867> ?p ?o ?g. }
Showing items 1 to 76 of
76
with 100 items per page.
- W4368754867 abstract "Recent advances in using language models to obtain cross-modal audio-text representations have overcome the limitations of conventional training approaches that use predefined labels. This has allowed the community to make progress in tasks like zero-shot classification, which would otherwise not be possible. However, learning such representations requires a large amount of human-annotated audio-text pairs. In this paper, we study unsupervised approaches to improve the learning framework of such representations with unpaired text and audio. We explore domain-unspecific and domain-specific curation methods to create audio-text pairs that we use to further improve the model. We also show that when domain-specific curation is used in conjunction with a soft-labeled contrastive loss, we are able to obtain significant improvement in terms of zero-shot classification performance on downstream sound event classification or acoustic scene classification tasks." @default.
- W4368754867 created "2023-05-05" @default.
- W4368754867 creator A5000750647 @default.
- W4368754867 creator A5009630456 @default.
- W4368754867 creator A5023830739 @default.
- W4368754867 creator A5029594137 @default.
- W4368754867 creator A5038903729 @default.
- W4368754867 creator A5042472021 @default.
- W4368754867 creator A5057729011 @default.
- W4368754867 date "2023-05-02" @default.
- W4368754867 modified "2023-09-28" @default.
- W4368754867 title "Unsupervised Improvement of Audio-Text Cross-Modal Representations" @default.
- W4368754867 doi "https://doi.org/10.48550/arxiv.2305.01864" @default.
- W4368754867 hasPublicationYear "2023" @default.
- W4368754867 type Work @default.
- W4368754867 citedByCount "0" @default.
- W4368754867 crossrefType "posted-content" @default.
- W4368754867 hasAuthorship W4368754867A5000750647 @default.
- W4368754867 hasAuthorship W4368754867A5009630456 @default.
- W4368754867 hasAuthorship W4368754867A5023830739 @default.
- W4368754867 hasAuthorship W4368754867A5029594137 @default.
- W4368754867 hasAuthorship W4368754867A5038903729 @default.
- W4368754867 hasAuthorship W4368754867A5042472021 @default.
- W4368754867 hasAuthorship W4368754867A5057729011 @default.
- W4368754867 hasBestOaLocation W43687548671 @default.
- W4368754867 hasConcept C121332964 @default.
- W4368754867 hasConcept C1276947 @default.
- W4368754867 hasConcept C134306372 @default.
- W4368754867 hasConcept C154945302 @default.
- W4368754867 hasConcept C185592680 @default.
- W4368754867 hasConcept C188027245 @default.
- W4368754867 hasConcept C204321447 @default.
- W4368754867 hasConcept C2779662365 @default.
- W4368754867 hasConcept C28490314 @default.
- W4368754867 hasConcept C3017588708 @default.
- W4368754867 hasConcept C33923547 @default.
- W4368754867 hasConcept C36503486 @default.
- W4368754867 hasConcept C41008148 @default.
- W4368754867 hasConcept C49774154 @default.
- W4368754867 hasConcept C59656382 @default.
- W4368754867 hasConcept C62520636 @default.
- W4368754867 hasConcept C71139939 @default.
- W4368754867 hasConceptScore W4368754867C121332964 @default.
- W4368754867 hasConceptScore W4368754867C1276947 @default.
- W4368754867 hasConceptScore W4368754867C134306372 @default.
- W4368754867 hasConceptScore W4368754867C154945302 @default.
- W4368754867 hasConceptScore W4368754867C185592680 @default.
- W4368754867 hasConceptScore W4368754867C188027245 @default.
- W4368754867 hasConceptScore W4368754867C204321447 @default.
- W4368754867 hasConceptScore W4368754867C2779662365 @default.
- W4368754867 hasConceptScore W4368754867C28490314 @default.
- W4368754867 hasConceptScore W4368754867C3017588708 @default.
- W4368754867 hasConceptScore W4368754867C33923547 @default.
- W4368754867 hasConceptScore W4368754867C36503486 @default.
- W4368754867 hasConceptScore W4368754867C41008148 @default.
- W4368754867 hasConceptScore W4368754867C49774154 @default.
- W4368754867 hasConceptScore W4368754867C59656382 @default.
- W4368754867 hasConceptScore W4368754867C62520636 @default.
- W4368754867 hasConceptScore W4368754867C71139939 @default.
- W4368754867 hasLocation W43687548671 @default.
- W4368754867 hasLocation W43687548672 @default.
- W4368754867 hasOpenAccess W4368754867 @default.
- W4368754867 hasPrimaryLocation W43687548671 @default.
- W4368754867 hasRelatedWork W1512718085 @default.
- W4368754867 hasRelatedWork W1569841287 @default.
- W4368754867 hasRelatedWork W2083892355 @default.
- W4368754867 hasRelatedWork W2293457016 @default.
- W4368754867 hasRelatedWork W2359001871 @default.
- W4368754867 hasRelatedWork W2369308426 @default.
- W4368754867 hasRelatedWork W2789919619 @default.
- W4368754867 hasRelatedWork W4214912084 @default.
- W4368754867 hasRelatedWork W1551406738 @default.
- W4368754867 hasRelatedWork W2610387714 @default.
- W4368754867 isParatext "false" @default.
- W4368754867 isRetracted "false" @default.
- W4368754867 workType "article" @default.