CosyVoice3 TTS-värden
Alibaba FunAudioLLM's latest multilingual model with ~150ms bi-streaming, instruction control, and zero-shot cloning.
Radera din text i SSML-taggar för exakt kontroll:
<speak><prosody rate="slow">Slow speech</prosody></speak>
Taggar den valda modellen förstår — klicka för att släppa en i din text där det händer:
Den här modellen läser vanlig text, så inline-taggar ignoreras. För taggbaserade känslor, byt till en uttrycksfull modell som Orpheus eller Bark.
Definiera egna uttal (ord = uttal):
Om jag inte kan 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.
Bäst för: Multilingual production TTS, real-time applications, voice cloning
Bläddra alla CosyVoice3 rösterMed en blick
- Utvecklare
- Alibaba (FunAudioLLM)
- Licens
- Apache 2.0
- Nivå
- standard
- Varvtal
- fast
- Röstkloning
- Ja, det är jag.
- Språk
- English, Chinese, Japanese, Korean, German, Spanish, French, Italian, Russian
- Max tecken
- 5000