Infinitetalk

Infinitetalk

Playground

Try it on WaveSpeedAI!

InfiniteTalk converts one photo + audio into audio-driven talking or singing avatar videos (Image-to-Video), up to 10 minutes, 720p tier $0.30/5s. Ready-to-use REST API, no coldstarts, affordable pricing.

Features

What is InfiniteTalk?

InfiniteTalk produces videos with precise lip sync, aligning the head, face, and body movements to the audio. It maintains identity across unlimited-length videos and also offers image-to-video generation, turning static photos into lively speaking or singing videos.

Why it looks great

  • Accurate lip synchronization: aligns lip motion precisely with audio, preserving natural rhythm and pronunciation.

  • Full-body coherence: captures head movements, facial expressions, and posture changes beyond the lips.

  • Identity preservation: maintains consistent facial identity and visual style across frames.

  • Image-to-video capability: turns static photos into realistic speaking or singing videos.

  • Instruction following: accepts text prompts to control scene, pose, or behavior while syncing to audio.

Pricing

Output ResolutionCost per 5 secondsMax Length
480p$0.1510 minutes
720p$0.3010 minutes

Billing Rules

  • Standard Rate: $0.03 per second
  • HD (720p) Rate: $0.06 per second (Double the Standard Rate)
  • Minimum Charge: All videos are billed for a minimum of 5 seconds (costing at least $0.15).
  • Billing Cap: To keep your costs predictable, billing is capped at 600 seconds (10 minutes).

How to Use

  1. Upload the audio file.
  2. Upload the image (the person to animate).
  3. (Optional) Paint a mask_image to specify which regions can move.
  4. (Optional) Add a prompt to guide expression, style, or pose.
  5. Select the resolution (480p or 720p).
  6. Set the seed (set a fixed number for reproducibility).
  7. Submit the job and download the result once it’s ready.

Note

  • Max clip length per job: up to 10 minutes
  • Processing speed: approximately 10–30 seconds of wall time per 1 second of video (varies by resolution and queue load)
  • Prompt: Avoid writing too many prompts and the recommend language is English, otherwise the model will make analysis errors, resulting in noisy and unexpect videos.

Mask safety tip:

Do not upload the full image as mask_image. The mask should only cover the regions you want to animate—otherwise the result may render as fully black.

More Versions

Reference

Authentication

For authentication details, please refer to the Authentication Guide.

API Endpoints

Submit Task & Query Result

set -euo pipefail

export WAVESPEED_API_KEY="your-api-key"

REQUEST_BODY=$(cat <<'JSON'
{
  "image": "https://interactive-examples.mdn.mozilla.net/media/cc0-images/painted-hand-298-332.jpg",
  "audio": "https://interactive-examples.mdn.mozilla.net/media/cc0-audio/t-rex-roar.mp3",
  "resolution": "480p",
  "seed": -1
}
JSON
)

# 1. Submit the prediction.
SUBMIT_RESPONSE=$(curl --silent --show-error --fail-with-body \
  -X POST "https://api.wavespeed.ai/api/v3/wavespeed-ai/infinitetalk" \
  -H "Authorization: Bearer ${WAVESPEED_API_KEY}" \
  -H "Content-Type: application/json" \
  -d "${REQUEST_BODY}")

TASK=$(printf '%s' "${SUBMIT_RESPONSE}" | jq 'if type == "object" and has("data") then .data else . end')
PREDICTION_ID=$(printf '%s' "${TASK}" | jq -r '.id // empty')
if [ -z "${PREDICTION_ID}" ]; then
  printf 'Submission response did not contain a prediction id
' >&2
  exit 1
fi
RESULT_URL=$(printf '%s' "${TASK}" | jq -r '.urls.get // empty')
if [ -z "${RESULT_URL}" ]; then RESULT_URL="https://api.wavespeed.ai/api/v3/predictions/${PREDICTION_ID}/result"; fi

# 2. Poll until the prediction finishes.
while true; do
  RESPONSE=$(curl --silent --show-error --fail-with-body \
    "${RESULT_URL}" \
    -H "Authorization: Bearer ${WAVESPEED_API_KEY}")
  RESULT=$(printf '%s' "${RESPONSE}" | jq 'if type == "object" and has("data") then .data else . end')
  STATUS=$(printf '%s' "${RESULT}" | jq -r '.status // empty')

  case "${STATUS}" in
    completed) printf '%s\n' "${RESULT}" | jq '.outputs'; break ;;
    failed|cancelled|timeout) printf '%s\n' "${RESULT}" | jq . >&2; exit 1 ;;
    created|processing) sleep 2 ;;
    *) printf 'Unexpected status: %s
' "${STATUS}" >&2; exit 1 ;;
  esac
done

Parameters

Task Submission Parameters

Request Parameters

ParameterTypeRequiredDefaultRangeDescription
imagestringYes-The image for generating the output.
audiostringYes--The audio for generating the output.
mask_imagestringNo-Optional mask image to specify the person in the image to animate.
promptstringNo-The positive prompt for the generation.
resolutionstringNo480p480p, 720pThe resolution of the output video.
seedintegerNo-1-The random seed to use for the generation. -1 means a random seed will be used.

Response Parameters

ParameterTypeDescription
codeintegerHTTP status code (e.g., 200 for success)
messagestringStatus message (e.g., “success”)
data.idstringUnique identifier for the prediction, Task Id
data.modelstringModel ID used for the prediction
data.outputsarrayOutput values, usually URL strings; some models return text strings or structured result objects (empty when status is not completed)
data.urlsobjectObject containing related API endpoints
data.urls.getstringURL to retrieve the prediction result
data.statusstringStatus of the task: created, processing, completed, or failed
data.created_atstringISO timestamp of when the request was created (e.g., “2023-04-01T12:34:56.789Z”)
data.errorstringError message (empty if no error occurred)
data.timingsobjectObject containing timing details
data.timings.inferenceintegerInference time in milliseconds

Result Request Parameters

ParameterTypeRequiredDefaultDescription
idstringYes-Task ID

Result Response Parameters

ParameterTypeDescription
codeintegerHTTP status code (e.g., 200 for success)
messagestringStatus message (e.g., “success”)
dataobjectThe prediction data object containing all details
data.idstringUnique identifier for the prediction
data.modelstringModel ID used for the prediction
data.outputsarray<string | object>Array of generated outputs (empty when status is not completed). Items are usually URL strings, but may be text strings or structured result objects, depending on the model.
data.urlsobjectObject containing related API endpoints
data.urls.getstringURL to poll for the prediction result
data.statusstringStatus: created, processing, completed, or failed
data.created_atstringISO timestamp of when the request was created
data.errorstringError message (empty if no error occurred)
data.timingsobjectObject containing timing details
data.timings.inferenceintegerInference time in milliseconds
© 2026 WaveSpeedAI. All rights reserved.