sleeping-ai/Rdiffusion

Rdiffusion数据集包含了Riffusion平台用户社区生成和分享的2000多首公开可用歌曲。该数据集提供了音频和视觉资源、详细的元数据（包括歌曲的ID、标题、作者ID、创建时间、时长、组ID等）、模型信息、使用标志、参与度信号（如播放次数、收藏次数）、歌词及时间戳等信息。所有音频文件都已准备好，可以直接用于机器学习和音乐生成研究。数据集在遵守欧洲研究豁免法的前提下，仅限于非商业研究使用。

The Rdiffusion dataset includes over 2000 publicly available songs generated and shared by the user community on the Riffusion platform. The dataset provides audio and visual assets, detailed metadata (including song ID, title, author ID, creation time, duration, group ID, etc.), model information, usage flags, engagement signals (such as play count, favorite count), lyrics and timestamps, and more. All audio files are ready for direct use in machine learning and music generation research. The dataset is legally scraped under European research exemption laws and is intended for non-commercial research use only.