LiveInfiniteTalk · audio-driven video generation2 endpoints

InfiniteTalk API — Audio-Driven Image & Video Lipsync

The InfiniteTalk API hub brings image-to-video and video-to-video lipsync into one Muapi catalog. Drive a portrait or existing clip with an audio URL, choose 480p or 720p, and keep one asynchronous request-and-poll workflow.

Get an InfiniteTalk API Key Try InfiniteTalk Image to Video Browse Lipsync Models

InfiniteTalk Image to VideoAnimate a portrait with speech and audio1 models

Audio to VideoLiveFeatured

Image to Video

InfiniteTalk Image to Video

Animate a still portrait with audio-driven lip movement, facial expressions, and subtle head gestures for a natural talking video.

480p / 720p

$0.20–$0.40/5 sec

Try Model

InfiniteTalk Video to VideoRe-sync an existing video to new audio1 models

Video to VideoLive

Video to Video

InfiniteTalk Video to Video

Reanimate an existing video to match new dialogue or speech while preserving the speaker's facial motion and expressions.

480p / 720p

$0.20–$0.40/5 sec

Try Model

What is the InfiniteTalk API?

InfiniteTalk is an audio-driven video family on Muapi. The Image to Video endpoint animates a still image with speech audio, while Video to Video reanimates an existing clip to match new dialogue or speech.

Both endpoints require an audio URL and a source image or video, accept an optional prompt, and support 480p or 720p output. Billing is based on audio duration in five-second blocks: 480p is $0.20 per five seconds and 720p is $0.40 per five seconds.

InfiniteTalk API capabilities

Portrait-to-talking-video

Animate a still portrait with an audio URL and optional prompt.

Video re-dubbing

Re-sync an existing video to new dialogue while retaining the original facial performance.

480p and 720p output

Choose the output resolution for the balance of cost and quality.

Duration-based billing

Pay from the audio duration in five-second blocks, with 720p charged at twice the 480p rate.

InfiniteTalk workflows

AI avatar messages

Turn a portrait and recorded voice into a personalized talking-head video.

Dubbing and localization

Use Video to Video with a translated audio track to re-sync existing dialogue.

Social and marketing clips

Create narrated portrait clips or update an existing campaign video with new audio.

Quick start — InfiniteTalk API code examples

Submit an InfiniteTalk image-to-video task with your Muapi key, receive a request ID, and poll the shared predictions endpoint for the completed video URL.

1. Submit an InfiniteTalk job

curl -X POST https://api.muapi.ai/api/v1/infinitetalk-image-to-video \
  -H "x-api-key: YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"image_url":"https://example.com/portrait.jpg","audio_url":"https://example.com/dialogue.wav","resolution":"480p","prompt":"Natural, friendly delivery"}'

# Response: {"request_id":"REQUEST_ID"}

The example calls Image to Video with image_url, audio_url, resolution, and an optional prompt. Switch to infinitetalk-video-to-video and use video_url for an existing clip.

2. Poll the result

curl https://api.muapi.ai/api/v1/predictions/REQUEST_ID/result \
  -H "x-api-key: YOUR_API_KEY"

Poll until the task completes, then read the generated video URL from the result response.

InfiniteTalk API FAQ

Which InfiniteTalk endpoints are available?

Muapi lists infinitetalk-image-to-video for portrait animation and infinitetalk-video-to-video for re-syncing an existing clip.

What inputs does InfiniteTalk require?

Image to Video requires image_url and audio_url; Video to Video requires video_url and audio_url. Both accept an optional prompt and resolution.

Which resolutions are supported?

Both endpoints support 480p and 720p, with 480p as the schema default.

How is InfiniteTalk priced?

Billing uses the supplied audio duration in five-second blocks: $0.20 per five seconds at 480p and $0.40 per five seconds at 720p.

Can I use InfiniteTalk for dubbing?

Yes. Provide the existing video in video_url and a new audio track to re-sync the speaker's mouth and expressions to the replacement dialogue.

Compare AI avatar and lipsync APIs →Compare InfiniteTalk with Muapi's other avatar and lipsync workflows.Compare AI video editing APIs →See InfiniteTalk video-to-video re-sync alongside other video editing endpoints.Browse lipsync models →Compare audio-driven video endpoints.Browse video editing models →Compare Video to Video workflows.Read the API docs →Learn Muapi's authentication, submit, and polling flow.

Build with InfiniteTalk

Connect portrait animation and video re-sync through one Muapi API key and a shared asynchronous workflow.

Get an API Key Try InfiniteTalk