Skip to main content

ByteDance Lip Sync is ByteDance's video lip-sync service, part of its video translation tools, with Lite for frontal single-speaker footage and Basic for complex single-speaker scenes. It returns audio-timed MP4 at 25 fps.

From $0.048/s

Workbench

Input

Lip sync
1/1

Drop a video here, or browse · 1 file · Max 90 MB · MP4 · WEBM

  • 1
1 / 1 files
Required in Lite and Basic, with no default: provide MOV, MP4, or MKV at 360p–1080p, 24–60 fps, and 1–30 Mbps. H.264 is recommended. Above-1080p video is reduced to 1080p and below-360p input is rejected. Its length does not set the billed seconds. at least 1 characters
affects price1/1
1 / 1 files
Required for either profile, with no default: clean human speech in MP3, WAV, AAC, M4A, or OGG, up to 10 MB. Duration must exceed 1 second and stay within 240 seconds for Lite or 210 for Basic. This duration sets billed seconds after upward rounding and determines the intended output length. at least 1 characters
Safety checkerOff

SpicyAPI does not run a safety checker on this model. Your prompt and media go to the model as written, and the model itself decides what comes back.

Checking your account before enabling submission and pricing.

Output

Ready for a new run
Example result

Related models and next tasks

Missing the model you need?

BYTEDANCE · VIDEO DIALOGUE

Choose a scene mode, supply the voice track, and let its timing set the edit.

Key numbers

Lite and Basic profiles
2 modes
Lite is for one frontal speaker; Basic handles complex single-speaker footage and can detect scenes.
Lite audio limit
240 s
Lite accepts up to 240 seconds of audio; Basic accepts up to 210. Both require more than 1 second.
Returned frame rate
25 fps
The completed MP4 uses 25 fps, including when the submitted footage uses another accepted frame rate.
Max source resolution
1080p
Footage above 1080p is reduced to 1080p; material below 360p is rejected.
Billed by audio seconds
Per sec
Round the uploaded recording up to whole seconds; Lite and Basic have the same per-second rate.
Lite loop directions
2 loops
Enable loop_reverse together with loop_video to alternate playback direction across repetitions.

Before you spend

All modes
  • Obtain rights to the faces, video, and voices you submit; every person shown must be an adult. Impersonation is prohibited.
  • Generated files are kept for 14 days, then deleted. Download the completed output before that retention window expires.
Lip sync
  • Both profiles target a single speaker. Lite expects frontal footage; Basic supports more complex scenes and optional scene detection.
  • Supply clean human speech longer than 1 second and no larger than 10 MB. The upper audio limit is 240 seconds in Lite or 210 seconds in Basic.
  • Use 360p–1080p source video at 24–60 fps and 1–30 Mbps. Higher-resolution footage is reduced to 1080p; below-360p material is not accepted.
  • Keep scene_detection for Basic. Omit noise removal, looping, reverse looping, and the start-position field from Basic requests; those controls belong to Lite.

What it does

RECORD-STORE EDIT

A shorter line trims the listening-room shot

In our tests, the record-store take ran 6.04 seconds before receiving a 5.2-second recording in Lite mode. The output stopped at 5.20 seconds, with the woman's mouth following the replacement line. Audio determined the ending of the edit.

MODE-SPECIFIC CONTROLS

Choose the controls your scene needs

Basic exposes scene_detection for shot boundaries and speaker recognition in single-speaker footage. Lite provides background-noise removal, looping, reverse looping, and a source start position. Mode-incompatible combinations are rejected when the request is accepted, before processing begins.

LONGER RECORDINGS

Repeat a take to finish the dialogue

Our Lite control trial combined loop_video, loop_reverse, video_start_seconds set to 1, and background-noise removal. A 5.04-second source covered a 6.5-second recording and returned 6.52 seconds of video. Review the repeated movement and its change of direction in your own edit.

MEASURED DELIVERY

Inspect the picture at 25 frames a second

The listening-room result measured 864×496 at 25 fps. The woman wears a burgundy satin camisole beneath a dark jacket, with a short dark pixie cut. This sample demonstrates revised speech timing in existing footage; its dimensions describe that particular output.

Where it fits

  • Replace a listening-room invitation

    Use the record-store example as a timing reference for an authorized dialogue revision. A shorter replacement recording can remove the unused ending while the mouth delivers the new words.

  • Dub a frontal presenter take

    Choose Lite for a single presenter facing the camera. Finish the translated or revised speech separately, then submit it with the source video for mouth alignment.

  • Process a single-speaker sequence with cuts

    Choose Basic and enable scene_detection when a single-speaker clip contains scene changes. Inspect synchronization around each cut after the asynchronous task completes.

  • Cover a longer voice-over with short footage

    Lite can loop the source when speech outlasts it. Set the starting second and consider alternating playback direction, then check whether the repeated action fits the intended delivery.

  • Prepare a voice track with background sound

    For a Lite job, remove_background_noise separates background sound from the driving voice before lip alignment. Listen to the source recording and review the returned speech; Basic requests require that control to be omitted.

Model reference

ByteDance Lip Sync API reference

ByteDance Lip Sync API specifications

Publisher / service
ByteDance Lip Sync, part of ByteDance's video translation tools
Request model
bytedance/lip-sync/lip-sync
Profile default
generation_mode: lite; basic also available
Video files
MOV, MP4, MKV; H.264 recommended
Source picture
360p–1080p; higher input reduced to 1080p
Source encoding range
24–60 fps; 1–30 Mbps
Speech files
MP3, WAV, AAC, M4A, OGG; up to 10 MB
Speech duration
More than 1 s; Lite up to 240 s, Basic up to 210 s
Result timing and format
Audio-led duration; MP4 at 25 fps
Pricing rule
Rounded-up audio seconds; equal rate for Lite and Basic
Optional scene analysis
scene_detection in Basic; off by default
Result availability
14 days, followed by deletion

How to call the ByteDance Lip Sync API

  1. Inspect the source and recording

    Check that the video meets the resolution, frame-rate, and bitrate requirements. Prepare clean speech in an accepted audio format, then measure its duration and file size.

  2. Select Lite or Basic

    Use Lite for frontal single-speaker video or Basic for complex single-speaker scenes. Check the audio ceiling for that mode before adding optional controls.

  3. Submit a compatible input set

    Provide video_url and audio_url to bytedance/lip-sync/lip-sync with generation_mode if needed. Add Basic scene detection or the relevant Lite controls, then retain the returned task identifier.

  4. Review timing and save the result

    Poll the task or receive its completion callback, then inspect the MP4 for speech alignment, cuts, and repeated footage. Download the approved edit within 14 days.

The short version

Let the recording define the length you need, then select the profile that fits your footage. The record-store edit shows a 6.04-second take becoming a 5.20-second line at 25 fps; the separate Lite loop trial shows short footage covering longer speech. Keep Basic scene analysis separate from Lite noise removal and loop settings. Measure the audio before submission, since its rounded-up seconds determine billing in either profile. After completion, check the spoken words, cut boundaries, and any repeated action, then download the MP4 before its 14-day retention period ends.

ByteDance Lip Sync API questions