taresh18/indic-speech
收藏Hugging Face2026-04-03 更新2026-04-12 收录
下载链接：
https://hf-mirror.com/datasets/taresh18/indic-speech
下载链接
链接失效反馈官方服务：
资源简介：
---
license: cc-by-4.0
language:
- en
- hi
task_categories:
- text-to-speech
size_categories:
- 10K<n<100K
---
# Indic Speech
Bilingual (English + Hindi) speech dataset for training Text-to-Speech and neural audio codec models. 24kHz mono audio.
| Split | Samples | Hours |
|-------|---------|-------|
| Train | 83,670 | 166.5 |
| Dev | 3,175 | 6.3 |
| Test | 939 | 2.0 |
**Total:** 87,784 samples, ~175 hours, 806 unique speakers.
## Source Datasets
| Dataset | Language |
|---------|----------|
| LibriTTS-R | English |
| IndicVoices-R | Hindi |
| SYSPIN | Hindi |
提供机构：
taresh18
