P4-01: LP-MusicCaps: LLM-Based Pseudo Music Captioning
Seungheon Doh (KAIST)*, Keunwoo Choi (Gaudio Lab, Inc.), Jongpil Lee (Neutune), Juhan Nam (KAIST)
Subjects (starting with primary): MIR fundamentals and methodology -> metadata, tags, linked data, and semantic web ; MIR fundamentals and methodology -> multimodality ; MIR tasks -> automatic classification
Presented In Person: 10-minute long-format presentation
Automatic music captioning, which generates natural language descriptions for given music tracks, holds significant potential for enhancing the understanding and organization of large volumes of musical data. Despite its importance, researchers face challenges due to the costly and time-consuming collection process of existing music-language datasets, which are limited in size. To address this data scarcity issue, we propose the use of large language models (LLMs) to artificially generate the description sentences from large-scale tag datasets. This results in approximately 2.2M captions paired with 0.5M audio clips. We term it Large Language Model based Pseudo music caption dataset, shortly, LP-MusicCaps. We conduct a systemic evaluation of the large-scale music captioning dataset with various quantitative evaluation metrics used in the field of natural language processing as well as human evaluation. In addition, we trained a transformer-based music captioning model with the dataset and evaluated it under zero-shot and transfer-learning settings. The results demonstrate that our proposed approach outperforms the supervised baseline model.
If the video does not load properly please use the direct link to video