Solos

Solos数据集是由西班牙巴塞罗那的庞培法布拉大学信息与通信技术系创建的，包含755个来自YouTube的独奏音乐表演视频，涵盖13种不同乐器。数据集通过人工筛选确保无背景噪音，提供高质量的音频和视频数据。创建过程中，使用OpenPose系统为每个视频帧提供身体和手部骨骼数据，并附带时间戳。Solos数据集主要应用于音频-视觉自我监督学习任务，如源分离、声音定位、跨模态生成和音频-视觉对应关系研究，旨在解决音乐信息检索中的多模态问题。

The Solos dataset was created by the Department of Information and Communication Technologies, Pompeu Fabra University, Barcelona, Spain. It contains 755 solo music performance videos sourced from YouTube, covering 13 distinct musical instruments. The dataset has been manually screened to eliminate background noise and provides high-quality audio and video data. During its construction, the OpenPose system was employed to generate body and hand skeletal data for each video frame, along with corresponding timestamps. The Solos dataset is primarily applied to audio-visual self-supervised learning tasks, including source separation, sound localization, cross-modal generation, and audio-visual correspondence research, aiming to address multimodal problems in music information retrieval.