Feng
Cartesia · taiwanese-mandarin · masculine
Precise Taiwanese Mandarin voice for service queries and updates.
AI video generator · Chinese
Yes. Seedance 2.5 and MiniMax H3 Max generate spoken Chinese dialogue together with the picture, so you can ask ChatGPT or Claude for a Chinese video in one request. When you need an exact voice, generate the line with one of 45 native Chinese voices and lip-sync it.
100 free credits · no subscription · estimate your cost
| Model | Chinese speech |
|---|---|
| Seedance 2.5 | Native dialogue, generated with the picture |
| MiniMax H3 Max | Native dialogue, generated with the picture |
| Gemini Omni Flash | Not documented, results may vary |
| MiniMax H3 Max Lip Sync | Lip-syncs the Chinese speech you supply |
| HeyGen Avatar 4 | Lip-syncs the Chinese speech you supply |
From provider documentation, checked 6 October 2026.
Play a voice, then name it in your prompt. See all 45 Chinese voices.
Cartesia · taiwanese-mandarin · masculine
Precise Taiwanese Mandarin voice for service queries and updates.
Cartesia · mandarin · feminine
Clear Mandarin voice for reliable business communication.
ElevenLabs · standard · masculine
A middle-aged Chinese voice in a calm, friendly tone. Perfect for narrative and storytelling.
ElevenLabs · beijing mandarin · feminine
A youthful Chinese female voice with lively energy. Suitable for narrating stories and creating content for social media videos.
HeyGen · masculine
James Gao
Fish Audio · masculine
This is a clear and professional female voice with a youthful and informative tone, well-suited for commercial announcements and broadcasting. The delivery is crisp and confident, making it ideal for retail or public address settings.
Portraits are AI-generated illustrations, not likenesses of the voice actors.
Add it from the ChatGPT plugin directory or at claude.ai/customize/plugins, and sign in. No API keys.
Say who is on screen and what they say. Seedance 2.5 and MiniMax H3 Max render the Chinese speech with the picture, so one request returns a clip with sound.
Pick one of 45 Chinese voices, generate the line as audio, then lip-sync it onto your presenter with MiniMax H3 Max Lip Sync. The recording drives the mouth and stays as the soundtrack.
Creative Claw shows the credit estimate before each generation. Download the MP4 when it is done.
Try this prompt
Use Creative Claw to make a 10-second vertical video with Seedance 2.5: a shop owner looks at the camera and welcomes new customers in Chinese. Write the Chinese line yourself and show it to me with the credit estimate before you generate.
Opens a new chat with the prompt ready. If Creative Claw is not installed yet, the chat offers to add it.
Yes. Seedance 2.5 and MiniMax H3 Max generate spoken Chinese dialogue together with the picture, so you can ask ChatGPT or Claude for a Chinese video in one request. When you need an exact voice, generate the line with one of 45 native Chinese voices and lip-sync it. Connect Creative Claw first.
Seedance 2.5 and MiniMax H3 Max are documented for native Chinese dialogue. Gemini Omni Flash is not documented for Chinese, so results may vary. MiniMax H3 Max Lip Sync and HeyGen Avatar 4 lip-sync any speech you supply, whatever the language.
Yes. Clone your voice from a short recording, generate the Chinese line with it, and lip-sync that audio onto the presenter. How well a clone speaks Chinese depends on the speech model, so test a short line first.
It depends on the model and the length. Seedance 2.5 is Usage-based; ask for an estimate (5s upfront estimates: 111 credits at 480p, 240 at 720p, 587 at 1080p. Text only, 16:9, native audio, no references. Final charge follows actual usage and cannot exceed the reservation.). Speech with ElevenLabs v4 is 20 credits / 1,000 chars. $10 buys 1,000 credits, there is no subscription, and new accounts get 100 free credits.