Hi-Fi TTS
292-hour high-fidelity multi-speaker English TTS dataset with 10 speakers each exceeding 17 hours at 44.1 kHz
Duration
292 hours
Languages
1
Sample Rate
44.1 kHz
Published
2021-04
Description
1Based on LibriVox audiobooks and Project Gutenberg texts
2Approximately 292 hours of speech with 10 speakers each contributing at least 17 hours
344.1 kHz high sampling rate
4Designed for high-quality multi-speaker TTS training
Language Details
| Language | Duration |
|---|---|
| English | 292 hours |
Publisher