Opencpop

High-quality Mandarin singing-voice synthesis corpus of 100 popular Chinese pop songs (3756 utterances) sung by a single female professional vocalist, 44.1 kHz, with phoneme/note boundary and pitch annotations.

• Task: tts
• Languages: zh
• Hours: 5.2
• Domain: mandarin pop singing
• License: CC BY-NC-ND 4.0
• Homepage: https://wenet-e2e.github.io/opencpop/
• Paper: https://arxiv.org/abs/2201.07429

Recommendation

Default open benchmark for Mandarin singing-voice synthesis (SVS) — pick for SVS prototyping, F0/note-conditional TTS, and karaoke experiments. Single-singer female vocal limits speaker generalization, and the CC BY-NC-ND license blocks commercial use and derivative redistribution.

Getting the data

Obtain from the dataset homepage.

Non-commercial, no-derivatives license; single-singer corpus.

Suggested processing

A recommended VoxKitchen pipeline ships in the repository at voxkitchen/templates/pipelines/tts-data-prep.yaml — run it with vkit docker run.