- Metadata: Please download our whole dataset at CapSpeech. If you want to use the detailed traits, please refer to our subsets: CapSpeech-PT, CapSpeech-PT-SEDB-HQ, CapTTS-SFT, CapSpeech-SEDB, CapSpeech-SEDB-test and CapSpeech-AgentDB.
- Open-source audio files: Please download the open-sourced audios from their official websites: Emilia-EN, GigaSpeech, CommonVoice, MLS-English, LibriTTS-R, VoxCeleb and VoxCeleb2, EARS, Expresso, VCTK.
- Our new collected audio files: Please download our CapSpeech-PT-SEDB, CapSpeech-PT-SEDB-HQ, CapSpeech-SEDB, and CapSpeech-AgentDB.