VidTL - The Editor That Edits For You

Noise Suppression

Reduce steady background noise and static locally in your browser.

Audio clips to process 0

Selected video clips use their linked audio when available.

Analyzing selected audio...

Suppression limit --

Auto-set from the selected clip. Higher values allow stronger noise reduction.

Post-filter smoothing --

Auto-set from the selected clip. Higher values can reduce leftover static.

Replace selected audio clips on the timeline

The processed WAV is also added to the media library.

Remove Silence (Local)

Detects and removes silence using waveform analysis.

Apply to all clips on timeline (including video clips)

When unchecked, only processes selected clip(s)

Only process talking-head clips (skip b-roll)

Uses subtitle density to detect talking-head clips and skip b-roll footage

Silence Threshold (dB) Auto

Auto

Min Silence Duration (seconds)

0.3

Pre-head (seconds)

0.0

Negative = cut into audio, Positive = leave padding

Post-tail (seconds)

0.0

Background Removal Options

Lower sensitivity keeps more of the subject when the mask removes too much.

Sensitivity

50%

Lower keeps more of the subject. Higher removes more uncertain pixels.

Edge feather

0.0px

Softens the transparent edge after the mask is created.

Background Removal Preview

Before

After

Generate Subtitles (Priority)

Select one or more clips on the timeline to generate subtitles with word-by-word timestamps.

Link subtitle clips to source clip in a group

Text to Speech

Convert text to speech using different AI models. Maximum 5000 characters.

Model

Voice

Quality Preset

Expression

Exaggeration 0.50

How emotional the voice sounds. 0.5 is neutral; higher feels more dramatic.

Pacing 0.50

Lower values slow down the pacing; higher values stick closer to the reference style.

Temperature 0.80

How varied the speech sounds across generations.

Seed

Use the same seed to repeat a result. 0 picks a new one each time.

Reference Audio

Optional. Upload a clear 6-10 second sample of the voice you want to clone.

Text to Synthesize *

0/5000 characters

Generate Music

Generate original music from a text description.

Description *

Describe the mood, instruments, tempo, and genre. 0/500

Model

Larger models take more time.

Length (seconds)

How long the generated music should be. Maximum 120 seconds.

Temperature: 1.0

Higher values produce more variety; lower values stay closer to the prompt.

Guidance: 3.0

How strongly the prompt steers the result. Higher = more faithful to the description.

Top-K: 250

Limits sampling to the K most likely choices. Set to 0 to disable.

Top-P: 0.0

Use nucleus sampling instead of Top-K. Set to 0 to disable.

Seed

Leave at -1 for random. Use a fixed number to reproduce a result.

Generate Video (AI)

Create a brand-new video clip from a text description. This is the slowest, most demanding option — try stock footage, your own uploads, and motion graphics first.

Description *

Describe the scene, subject, camera, and style. 0/2000

Avoid (optional)

Things you do not want to appear in the video.

Model

LTX 2.3 is currently available.

Seed

Leave at -1 for random. Use a fixed number to reproduce a result.

Width

Height

Frame rate

Length (frames)

Total frames. Seconds = frames ÷ frame rate.

Steps: 8

More steps can improve quality but take longer.

Guidance: 1

How strongly the description steers the result.

Denoise: 1

Keep at 1.0 for fully generated video.

Sampler

Scheduler

Starting Image (optional)

Optional. The video will animate from this image (image-to-video).

Generating video uses a large amount of your usage allowance and can take several minutes. You are only charged if generation succeeds.

Auto B-Roll (Pexels)

Uses your existing subtitle clips to find matching stock video and place it on the timeline.

Segment Length (seconds)

Max Segments

Results Per Segment

Fallback Query

Place clips on a new "B-Roll" track Use fallback query if no results

FFmpeg

Run custom FFmpeg commands on your media files.

Note: Input files can be any size, but output files are capped at 2GB to prevent browser issues.

Available Media Files

Click a file to insert its name into the command

FFmpeg Command

Filenames with spaces will be automatically quoted

Output Filename (optional)

Used if not specified in command

Quick Examples

Extract audio 10s clip @ 5s First 30s Remove audio Replace audio Scale to 720p

Tip: -ss before -i for faster/accurate seeking

Export Video

File Name

Format

Resolution

Quality

80

Higher quality = larger file size. For ProRes, this selects different quality levels.

Transparent background

Available for WebM exports. Empty canvas areas stay transparent.

Audio Channels

Use Mono if your exported video only has audio in one speaker

Video Settings

Video Codec

Frame Rate

Encoding Preset

Pixel Format

Rate Control

Tune (H.264 / H.265)

Keyframe Interval (seconds)

Audio Settings

Audio Codec

Audio Bitrate

Sample Rate

Channels

Normalize loudness (automatic volume balancing)

Select a platform preset to automatically configure optimal settings

Export Timeline (FCP 7 XML)

Compatible with Premiere Pro, Final Cut Pro 7, DaVinci Resolve

Export Options

XML only

XML with all media files (ZIP)

Clips with effects, masks, or text will be rendered as video when exporting with media.

About VidTL Editor

VidTL

A browser-based video editor.

Editor Paused

Keyboard Shortcuts

Noise Suppression

Remove Silence (Local)

Preview

Background Removal Options

Background Removal Preview

Before

After

Generate Subtitles (Priority)

Text to Speech

Generate Music

Generate Video (AI)

Auto B-Roll (Pexels)

FFmpeg

Export Video

Video Settings

Audio Settings

Export Timeline (FCP 7 XML)

About VidTL Editor