CosyVoice3 TTS
Alibaba FunAudioLLM's latest multilingual model with ~150ms bi-streaming, instruction control, and zero-shot cloning.
Omotajte tekst u SSML oznake za preciznu kontrolu:
<speak><prosody rate="slow">Slow speech</prosody></speak>
Oznake koje odabrani model razumije — kliknite da biste ih ubacili u tekst gdje se pojavljuju:
Ovaj model čita običan tekst, tako da se inline oznake ignoriraju. Za emocije zasnovane na oznakama, prebacite se na ekspresivni model poput Orpheusa ili Bark-a.
Definirajte vlastite izgovore (riječ = izgovor):
O meni CosyVoice3
CosyVoice3 is the newest generation from Alibaba's FunAudioLLM team and a clear step up from CosyVoice 2. It introduces bi-streaming inference with roughly 150ms latency and instruction-based control, letting you steer emotion, speed, and volume through prompts. Speaker similarity for zero-shot voice cloning is improved, and coverage spans 9 languages plus 18 Chinese dialects. An RL-tuned variant pushes prosody to a state-of-the-art level. With a 5,000-character ceiling, fast generation, and strong cloning, it's geared toward multilingual production TTS and real-time applications.
Najbolje za: Multilingual production TTS, real-time applications, voice cloning
Pregledaj sve CosyVoice3 glasoviNa prvi pogled
- Programer
- Alibaba (FunAudioLLM)
- Licenca
- Apache 2.0
- Životinje
- standard
- Brzina
- fast
- Kloniranje glasa
- Da.
- Jezici
- English, Chinese, Japanese, Korean, German, Spanish, French, Italian, Russian
- Maksimalan broj znakova
- 5000