AISTAT lab system for DCASE2025 Task6: Language-based audio retrieval
Fuente:
arXiv
Guardado en:
| Autores principales: | , , |
|---|---|
| Formato: | Preprint |
| Publicado: |
2025
|
| Materias: | |
| Acceso en línea: | |
| Etiquetas: |
Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
|
| _version_ | 1866916958991024128 |
|---|---|
| author | Kim, Hyun Jun Choi, Hyeong Yong Lim, Changwon |
| author_facet | Kim, Hyun Jun Choi, Hyeong Yong Lim, Changwon |
| contents | This report presents the AISTAT team's submission to the language-based audio retrieval task in DCASE 2025 Task 6. Our proposed system employs dual encoder architecture, where audio and text modalities are encoded separately, and their representations are aligned using contrastive learning. Drawing inspiration from methodologies of the previous year's challenge, we implemented a distillation approach and leveraged large language models (LLMs) for effective data augmentation techniques, including back-translation and LLM mix. Additionally, we incorporated clustering to introduce an auxiliary classification task for further finetuning. Our best single system achieved a mAP@16 of 46.62, while an ensemble of four systems reached a mAP@16 of 48.83 on the Clotho development test split. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2509_16649 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | AISTAT lab system for DCASE2025 Task6: Language-based audio retrieval Kim, Hyun Jun Choi, Hyeong Yong Lim, Changwon Sound Artificial Intelligence Audio and Speech Processing This report presents the AISTAT team's submission to the language-based audio retrieval task in DCASE 2025 Task 6. Our proposed system employs dual encoder architecture, where audio and text modalities are encoded separately, and their representations are aligned using contrastive learning. Drawing inspiration from methodologies of the previous year's challenge, we implemented a distillation approach and leveraged large language models (LLMs) for effective data augmentation techniques, including back-translation and LLM mix. Additionally, we incorporated clustering to introduce an auxiliary classification task for further finetuning. Our best single system achieved a mAP@16 of 46.62, while an ensemble of four systems reached a mAP@16 of 48.83 on the Clotho development test split. |
| title | AISTAT lab system for DCASE2025 Task6: Language-based audio retrieval |
| topic | Sound Artificial Intelligence Audio and Speech Processing |
| url | https://arxiv.org/abs/2509.16649 |