VCTK
Multi-speaker speech dataset with 109 native English speakers recorded at 48 kHz studio quality, widely used for multi-speaker TTS research
Duration
44 hours
Languages
1
Sample Rate
48 kHz
Published
2019-07
Description
1109 native English speakers each reading approximately 400 sentences
2Recorded in a professional recording studio with high audio quality
3Speakers from various English accent regions
4ODC-By v1.0 license
Language Details
| Language | Duration |
|---|---|
| English | 44 hours |
Publisher