Skip to content

Dataset · espnet

Chinese singing voice segments

espnet/ace-opencpop-segments

A large audio-text corpus for synthesis and recognition of Chinese vocals.

What's inside

A dataset from the ESPnet project (ACE-Opencpop) featuring from 100 thousand to 1 million singing voice audio segments and corresponding texts. The Chinese-language data is intended for speech and singing synthesis, audio conversion, and ASR tasks.

What you can do with it

  • Training Chinese singing voice synthesis (SVS) models
  • Training vocal recognition and voice conversion (audio-to-audio) systems
Tasks
Text → audio, Audio → audio, Automatic speech recognition
Languages
Chinese
Size
100K–1M rows
Downloads this month
1.5K
Likes
10

How to load it

from datasets import load_dataset

ds = load_dataset("espnet/ace-opencpop-segments")