Get a weekly rundown of the latest AI models and research... subscribe!

Video Retalking



The "video-retalking" model is designed for audio-based lip synchronization in talking head videos. The user provides links to a video file (containing a face) and an audio file as inputs. The model then synchronizes the lip movements in the video with the provided audio, creating the illusion that the person in the video is speaking the audio. The output is a link to the generated MP4 video file, where the face in the video has been modified to match the audio input.

Use cases

The AI model for video-retalking, as described by its creator, is a technology that utilizes audio-based lip synchronization for talking head videos. This can accommodate numerous potential applications. In film production and animation, it could be used to correct mismatches between audio dialogue and actor or animated character's lip movements, improving overall viewer experience. It opens up possibilities in creating dubbed content for foreign films, where the model can be used to tweak lip movements for synchronization with translated dialogues so that it appears as if the characters are speaking the translated language. Furthermore, this AI model could also be used in video conferencing interfaces, where network lags cause speech and video to be out-of-sync. Live broadcasters or vloggers can use this technology to improve their on-screen presence, making their videos more professional. In virtual reality (VR) applications, it enhances the realism of characters, making for a more immersive interaction with the VR environment. Tech firms could speculate upon creating software products embedded with this AI model for creators in the film, animation, VR industry, vloggers, and the generalized public who use video conferencing regularly. Educational platforms can utilize it to synchronize the voices of educators with their video, especially for language learning applications, where correct lip movements play a critical role. In essence, the video-retalking AI model presents promising possibilities in any domain that relies on video content production or consumption.


Cost per run
Avg run time

Creator Models

Dreambooth Pikachu$0.08195513
Night Enhancement$0.0104538,658

Similar Models

No similar models found

Try it!

You can use this area to play around with demo applications that incorporate the Video Retalking model. These demos are maintained and hosted externally by third-party creators. If you see an error, message me on Twitter.

Currently, there are no demos available for this model.


Summary of this model and related resources.

Model NameVideo Retalking
Audio-based Lip Synchronization for Talking Head Video
Model LinkView on Replicate
API SpecView on Replicate
Github LinkView on Github
Paper LinkView on Arxiv


How popular is this model, by number of runs? How popular is the creator, by the sum of all their runs?

Model Rank
Creator Rank


How much does it cost to run this model? How long, on average, does it take to complete a run?

Cost per Run$-
Prediction Hardware-
Average Completion Time-