140+ Languages Supported
140
Controllability and Security
Emotion Variations per Voice
defaultchatcustomerservicenarration-professionalnewscast-casualnewscast-formalcheerfulempatheticangrysadexcitedfriendlyterrifiedshoutingunfriendlywhisperinghopefulfearfulfunnyrelievedshyseriousassistantconversationnewscast
Fine-Tuning Speaking Style
Full Model Training Hours
100000
High-Fidelity Speech Synthesis
Merged Feature for Long Voiceovers
Open Source Model Training Hours
40000
Sample Rate for Audio Output
24000
Voice Customization Options