Alibaba Cloud · Text to speech
CosyVoice 3.5 Plus
Expressive speech synthesis with faithful voice cloning
Pricing
Charged by actual usage
Unit: Points
Text to speech
- Per 100 characters
- 5≈ ¥0.05Member price 3≈ ¥0.03
- Per 10,000 characters
- 500≈ ¥5Member price 300≈ ¥3
| Per 100 characters | Per 10,000 characters | |
|---|---|---|
| List price | 5≈ ¥0.05 | 500≈ ¥5 |
| Member price (40% off) | 3≈ ¥0.03 | 300≈ ¥3 |
- Paid members are charged the member price, 40% below the list price.
- Billed by the number of characters synthesized.
Lifetime and Team members get 40% off this model. See plans
About this model
CosyVoice 3.5 Plus is Tongyi Lab’s expressive speech synthesis model. It improves voice cloning and design while focusing on speaker similarity, pronunciation, prosody, and audio quality. Multilingual synthesis and style instructions support narration and personalized voices.
Use in Beav
- 01Download Beav and sign in
- 02Choose CosyVoice 3.5 Plus in the model picker
- 03Use it for turning scripts into natural voiceovers
Similar models
View allMiniMax
speech-2.8-turbo异步
Fast, natural, and expressive speech synthesis
8.33 points / 100 characters
Google
Gemini 3.8 Flash TTS
Expressive speech with nuanced vocal performance
Input 1,167 · Output 21,000 points / 1M tokens
Google
Gemini 3.8 Flash-Lite TTS
Efficient speech generation for everyday production
Input 1,167 · Output 14,000 points / 1M tokens
Google
gemini-3.1-flash-tts
Steerable multilingual speech generation preview
Input 2,917 · Output 58,333 points / 1M tokens
