ByteDance Latentsync | AI Digital Human API

LatentSync

LatentSync is an advanced AI-powered lip synchronization model that matches video lip movements to any audio track. Upload a video and audio file, and the model automatically adjusts the speaker's lips to perfectly sync with the new audio — ideal for dubbing, translation, and content localization.

Why It Stands Out

Precise lip synchronization: AI analyzes audio and adjusts lip movements frame by frame for natural results.
Latent-based approach: Advanced latent space manipulation for high-quality, realistic output.
Audio replacement: Sync any audio track to existing video footage seamlessly.
High-quality output: Produces realistic lip movements that blend naturally with the original video.
Simple workflow: Just upload audio and video — no manual editing required.

Parameters

Parameter	Required	Description
audio	Yes	Audio file to sync (upload or public URL).
video	Yes	Video file with the face to be synced (upload or public URL).

How to Use

Upload your audio — drag and drop a file or paste a public URL.
Upload your video — provide the video containing the face to be lip-synced.
Click Run and wait for processing.
Preview and download the lip-synced video.

Best Use Cases

Video Dubbing — Replace original audio with dubbed versions in different languages.
Content Localization — Adapt videos for international audiences with synced translations.
Film & TV Production — Fix audio sync issues or replace dialogue in post-production.
Marketing & Advertising — Create localized ad content without reshooting.
E-learning — Produce multilingual training videos from a single source.
Social Media — Create fun lip-sync content for entertainment.

Pricing

Duration	Price
5 seconds	$0.15
30 seconds	$0.90
60 seconds	$1.80

Billing Rules

Billed per 5 seconds of audio at $0.15
Minimum charge: 5 seconds
Maximum billable duration: 600 seconds (10 minutes)

Pro Tips for Best Quality

Use clear, high-quality audio for best lip-sync accuracy.
Ensure the face in the video is clearly visible and well-lit.
Front-facing videos with minimal head movement produce the best results.
Keep audio and video durations matched for optimal synchronization.
For best results, use videos where the speaker's mouth is clearly visible throughout.

Notes

Ensure uploaded audio and video URLs are publicly accessible.
Maximum billable duration is 600 seconds (10 minutes) per job.
Processing time varies based on video length and current queue load.
Please ensure your content complies with usage guidelines.

Latentsync API — Quick start

Grab a WaveSpeedAI API key, then call POST https://api.wavespeed.ai/api/v3/bytedance/latentsync with your input as JSON. The endpoint returns a prediction id. Start polling the result endpoint around every 2 seconds, increase the interval for long-running tasks, and stop on any terminal status. On completed, read URLs from data.outputs. Examples for Latentsync below.

HTTP example

# Submit the prediction
curl --fail-with-body --connect-timeout 10 --max-time 60 \
  -X POST "https://api.wavespeed.ai/api/v3/bytedance/latentsync" \
  -H "Content-Type: application/json" \
  -H "Authorization: Bearer $WAVESPEED_API_KEY" \
  -d '{
    "audio": "https://example.com/your-audio.mp3",
    "video": "https://example.com/your-input.mp4"
}'

# Wait at least 2 seconds, then poll. Safe GET requests may be retried.
curl --fail-with-body --connect-timeout 10 --max-time 30 \
  --retry 4 --retry-all-errors --retry-delay 1 \
  -X GET "https://api.wavespeed.ai/api/v3/predictions/{request_id}/result" \
  -H "Authorization: Bearer $WAVESPEED_API_KEY"

# Start at 2 seconds and increase the interval for long-running tasks.
# Stop on completed, failed, cancelled, or timeout.

Node.js example

// npm install wavespeed
const { Client } = require('wavespeed');

const apiKey = process.env.WAVESPEED_API_KEY;
if (!apiKey) throw new Error('Set WAVESPEED_API_KEY');
const client = new Client(apiKey);

try {
  const result = await client.run("bytedance/latentsync", {
        "audio": "https://example.com/your-audio.mp3",
        "video": "https://example.com/your-input.mp4"
}, {
    timeout: 3600,
    pollInterval: 2.0,
  });
  console.log(result.outputs);
} catch (error) {
  console.error('Generation failed:', error);
  process.exitCode = 1;
}

Python example

# pip install wavespeed
import os
from wavespeed import Client

client = Client(api_key=os.environ["WAVESPEED_API_KEY"])

try:
    output = client.run(
        "bytedance/latentsync",
        {
    "audio": "https://example.com/your-audio.mp3",
    "video": "https://example.com/your-input.mp4"
},
        timeout=3600.0,
        poll_interval=2.0,
    )
    print(output["outputs"])
except Exception as error:
    raise SystemExit(f"Generation failed: {error}") from error

Latentsync API — Frequently asked questions

What is the Latentsync API?

Latentsync is a ByteDance model for talking-avatar generation, exposed as a REST API on WaveSpeedAI. LatentSync combines Stable Diffusion and TREPA for high-res end-to-end lip-sync, delivering precise, realistic mouth motions in generated videos. Ready-to-use REST inference API, best performance, no coldstarts, affordable pricing. You can call it programmatically or try it from the playground above.

How do I call the Latentsync API?

POST your input parameters to the model's REST endpoint (shown in the API tab of this playground) with your WaveSpeedAI API key in the Authorization header. Submission returns a prediction ID. Poll the result endpoint starting around every 2 seconds, increase the interval for long-running tasks, and stop on any terminal status. The playground generates production-oriented Python, JavaScript, and cURL examples with timeouts, transient-error handling, and safe GET retries. Full request/response shape is documented at https://wavespeed.ai/docs/docs-api/bytedance/bytedance-latentsync.

How much does Latentsync cost per run?

Latentsync starts at $0.15 per run. That figure is the base price — the final charge scales with the parameters you set in the form (output size, length, count, references, or whatever knobs this model exposes), so a higher-quality or larger output costs more than a minimal one. The exact cost for your current input is shown live next to the Generate button before you submit, and the actual per-call charge is recorded on the prediction afterwards.

What inputs does Latentsync accept?

Key inputs: `video`, `audio`. The full JSON schema (types, defaults, allowed values) is rendered above the Generate button and mirrored in the API reference at https://wavespeed.ai/docs/docs-api/bytedance/bytedance-latentsync.

How long does Latentsync take to generate?

Average end-to-end generation time on WaveSpeedAI is around 121 seconds per request — measured across recent runs. Queue time scales with global demand; live status is visible in the prediction record.

Can I use Latentsync outputs commercially?

Commercial usage rights depend on the model's license, set by its provider (ByteDance). The license summary appears on the model card above; see WaveSpeedAI's Terms of Service for platform-level conditions.

ExamplesView all

Related Models

README