KlingAIResearch
X-Dub
Tool that redraws a video character's mouth to match a new audio track
- Wan 2.2 TI2V-5B
- Whisper large-v2
- Wav2Vec2-base-960h
- +2

Give InfiniteTalk a video or still photo plus audio, and it makes a new video where mouth, head and body follow the sound. Streaming mode keeps going past one minute for long dubs, though you'll need a big GPU and some setup. After about a minute, colors drift and the face matches less well.
Best for: Dubbing editors who want to re-animate existing footage or a portrait photo with matching lip sync.
From the InfiniteTalk README on GitHub. Tap an image to see it full size.
Rewritten from the README. Check the repo for the latest steps.
multitalk and install PyTorch 2.4.1 and xformers 0.0.28 with CUDA 12.1 wheels.pip install -r requirements.txtconda install -c conda-forge librosahuggingface-cli download.python generate_infinitetalk.py --ckpt_dir weights/Wan2.1-I2V-14B-480P --wav2vec_dir 'weights/chinese-wav2vec2-base' --infinitetalk_dir weights/InfiniteTalk/single/infinitetalk.safetensors --input_json examples/single_example_image.json --size infinitetalk-480 --sample_steps 40 --mode streaming --motion_frame 9 --save_file infinitetalk_respython app.py --ckpt_dir weights/Wan2.1-I2V-14B-480P --wav2vec_dir 'weights/chinese-wav2vec2-base' --infinitetalk_dir weights/InfiniteTalk/single/infinitetalk.safetensors --num_persistent_param_in_dit 0 --motion_frame 9Other editing & audio projects people compare with InfiniteTalk.
KlingAIResearch
Tool that redraws a video character's mouth to match a new audio track
bytedance
Re-syncs a talking-face video to another audio track
KlingAIResearch
Animates a face photo or video with motion from another clip
RVC-Boss
Clones a voice from a 5-second clip and reads your text in it
Do you maintain InfiniteTalk? Add this badge to your README so English speakers can find our write-up.
[](https://openmicrodrama.com/projects/meigen-ai-infinitetalk)The badge links to this page. Want your description changed? Email support@openmicrodrama.com.
Reviewed Oct 1, 2026. Last push 2026-05-22, outside our 90-day window. We kept it because it's still widely used. From the MeiGen-AI team; the same team has since released LongCat-Video-Avatar-1.5, an upgraded model. We write these descriptions ourselves. The repo's own docs are the final word.