Guardado en:
| Autores principales: | , , , , , , |
|---|---|
| Formato: | Preprint |
| Publicado: |
2026
|
| Materias: | |
| Acceso en línea: | https://arxiv.org/abs/2604.07895 |
| Etiquetas: |
Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
|
| _version_ | 1866910115566714880 |
|---|---|
| author | Shin, Joonhyeok Kang, Jaehoon Lee, Yujun Lee, Hannah Lee, Yejin Park, Yoonji Shim, Kyuhong |
| author_facet | Shin, Joonhyeok Kang, Jaehoon Lee, Yujun Lee, Hannah Lee, Yejin Park, Yoonji Shim, Kyuhong |
| contents | Selecting an appropriate background music (BGM) that supports natural human conversation is a common production step in media and interactive systems. In this paper, we introduce dialogue-conditioned BGM recommendation, where a model should select non-intrusive, fitting music for a multi-turn conversation that often contains no music descriptors. To study this novel problem, we present DialBGM, a benchmark of 1,200 open-domain daily dialogues, each paired with four candidate music clips and annotated with human preference rankings. Rankings are determined by background suitability criteria, including contextual relevance, non-intrusiveness, and consistency. We evaluate a wide range of open-source and proprietary models, including audio-language models and multimodal LLMs, and show that current models fall far short of human judgments; no model exceeds 35% Hit@1 when selecting the top-ranked clip. DialBGM provides a standardized benchmark for developing discourse-aware methods for BGM selection and for evaluating both retrieval-based and generative models. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2604_07895 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | DialBGM: A Benchmark for Background Music Recommendation from Everyday Multi-Turn Dialogues Shin, Joonhyeok Kang, Jaehoon Lee, Yujun Lee, Hannah Lee, Yejin Park, Yoonji Shim, Kyuhong Artificial Intelligence Selecting an appropriate background music (BGM) that supports natural human conversation is a common production step in media and interactive systems. In this paper, we introduce dialogue-conditioned BGM recommendation, where a model should select non-intrusive, fitting music for a multi-turn conversation that often contains no music descriptors. To study this novel problem, we present DialBGM, a benchmark of 1,200 open-domain daily dialogues, each paired with four candidate music clips and annotated with human preference rankings. Rankings are determined by background suitability criteria, including contextual relevance, non-intrusiveness, and consistency. We evaluate a wide range of open-source and proprietary models, including audio-language models and multimodal LLMs, and show that current models fall far short of human judgments; no model exceeds 35% Hit@1 when selecting the top-ranked clip. DialBGM provides a standardized benchmark for developing discourse-aware methods for BGM selection and for evaluating both retrieval-based and generative models. |
| title | DialBGM: A Benchmark for Background Music Recommendation from Everyday Multi-Turn Dialogues |
| topic | Artificial Intelligence |
| url | https://arxiv.org/abs/2604.07895 |