Domain adaptation method and modality gap impact in audio text models for prototypical sound classification
Mostra el registre complet Registre parcial de l'ítem
- dc.contributor.author Acevedo, Emiliano
- dc.contributor.author Rocamora, Martín
- dc.contributor.author Fuentes, Magdalena
- dc.date.accessioned 2025-09-04T06:41:34Z
- dc.date.available 2025-09-04T06:41:34Z
- dc.date.issued 2025
- dc.description.abstract Audio-text models are widely used in zero-shot environmental sound classification as they alleviate the need for annotated data. However, we show that their performance severely drops in the presence of background sound sources. Our analysis reveals that this degradation is primarily driven by SNR levels of background soundscapes, and independent of background type. To address this, we propose a novel method that quantifies and integrates the contribution of background sources into the classification process, improving performance without requiring model retraining. Our domain adaptation technique enhances accuracy across various backgrounds and SNR conditions. Moreover, we analyze the modality gap between audio and text embeddings, showing that narrowing this gap improves classification performance. The method generalizes effectively across state-of-the-art prototypical approaches, showcasing its scalability and robustness for diverse environments.
- dc.format.mimetype application/pdf
- dc.identifier.citation Acevedo E, Rocamora M, Fuentes M. Domain adaptation method and modality gap impact in audio text models for prototypical sound classification. In: Proceedings of Interspeech 2025; 2025 Aug 17-21; Rotterdam, Netherlands. [place unknown]: International Speech Communication Association; 2025. p. 1328-32. DOI: 10.21437/Interspeech.2025-886.
- dc.identifier.uri http://hdl.handle.net/10230/71110
- dc.language.iso eng
- dc.publisher International Speech Communication Association (ISCA)
- dc.relation.ispartof Proceedings of Interspeech 2025; 2025 Aug 17-21; Rotterdam, Netherlands. [place unknown]: International Speech Communication Association; 2025.
- dc.rights © 2025 ISCA
- dc.rights.accessRights info:eu-repo/semantics/openAccess
- dc.subject.keyword Audio text models
- dc.subject.keyword Sound classification
- dc.title Domain adaptation method and modality gap impact in audio text models for prototypical sound classification
- dc.type info:eu-repo/semantics/conferenceObject
- dc.type.version info:eu-repo/semantics/publishedVersion