Título:
Domain adaptation method and modality gap impact in audio-text models for prototypical sound classification.
Otros títulos:
Coordinador:
Director:
Compilador:
Autor:
Acevedo, Emiliano
Rocamora, Martín
Fuentes, Magdalena
Rocamora, Martín
Fuentes, Magdalena
Tutor:
Tipo de documento:
Ponencia
Editor:
Palabras clave:
Audio-text models
Modality gap
Domain adaptation
Zero-shot sound classification
Modality gap
Domain adaptation
Zero-shot sound classification
Descriptores:
Año de publicación:
2025
Contenido:
Resumen:
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.
Descripción:
metadata.articulos.dc.description.uri:
https://www.isca-archive.org/interspeech_2025/acevedo25_interspeech.html#
Editorial:
ISCA - International Speech Communication Association.
EN:
Interspeech 2025, Rotterdam, The Netherlands, 17-21 aug. 2025, pp. 1328-1332.
Financiadores:
Citación:
Acevedo, E., Rocamora, M. y Fuentes, M. Domain adaptation method and modality gap impact in audio-text models for prototypical sound classification [en línea]. EN: Interspeech 2025, Rotterdam, The Netherlands, 17-21 aug. 2025, pp. 1328-1332. DOI: 10.21437/Interspeech.2025-886.
Citación:
metadata.articulos.cc.license.name:
ISBN:
e-ISBN:
ISSN:
ISMN:
Otros identificadores:
Cobertura geográfica:
Cobertura temporal:
Departamento académico:
Procesamiento de Señales
Grupo de investigación:
Procesamiento de Audio (GPA)
Licencia:
Licencia Creative Commons Atribución - No Comercial - Sin Derivadas (CC - By-NC-ND 4.0)
Colecciones:
| Ficheros | Descripción | Tamaño | Formato | ||
|---|---|---|---|---|---|
| ARF25.pdf | Versión publicada | 405.38 KB | Adobe PDF |