Free Text-Based Video Editor

Select words in the transcript to cut the matching speech from video or audio. Correcting text updates captions and subtitles only.

Changing or correcting words only updates the captions and subtitles. It does not change the recorded speech or generate new audio. To remove speech from the recording, select the words and use Cut selected words; that removes the matching video or audio segment.

The first run downloads your chosen speech model and, if enabled, a separate speaker identification model. Your browser caches these downloads when storage allows, so later runs can start sooner. Switching to a model you have not used yet requires another download.

Yes. Enable Identify speakers before transcribing to label different voices in the transcript, such as Speaker 1 and Speaker 2. In Edit, right-click selected words to correct their speaker assignment, and rename labels to the names you want to show. You can use up to eight speaker labels and include or hide speaker names in captions and subtitle downloads. Automatic assignments may need review, especially when voices overlap.

Yes. In Clean up, choose Reduce background noise to process the recording locally. Once it is ready, toggle Use noise reduction to compare the cleaned audio with the original. Leave it enabled to apply the cleaned audio to your exported video or WAV file. Listen before exporting, since results depend on the recording.

After transcription, choose Remove filler words in Clean up to review suggestions for fillers such as um, uh, and erm. The editor also flags phrases such as you know and I mean for context review, plus possible filler sounds that may be missing from the transcript. Listen to suggestions, uncheck anything you want to keep, then remove the selected sections from the video or audio. These word and phrase rules focus on English fillers.

Yes. Choose Remove pauses in Clean up to find quiet sections. Pause settings let you choose how long a pause must be to count and how much silence to keep, so you can tighten the pacing while retaining breathing room. Review the suggestions and remove only the sections you want.

Yes. In Captions & export, add captions to your video with Outline, Boxed, Highlight words, or Simple styles. Adjust their size and place them at the top, middle, or bottom, with optional speaker names. You can also download SRT or VTT subtitles and a TXT transcript. Captions and downloads reflect your text corrections and omit speech you have cut.

Export an edited video as MP4 or WebM, or export just the audio as WAV. Video exports preserve the aspect ratio, up to 1920 pixels on the longest edge. Use Save directly to file to write the recording as it is created, which reduces the need to hold a large export in browser memory.

Use Chrome or Edge on a desktop or laptop. Mobile, Safari and Firefox editing are not supported.

No, media processing happens on your device. If you use an AI agent to help you with the edit, that AI agent will have access to your video.

Yes. Use the WebMCP-compatible browser AI assistant of your choice to help edit your recording. The editor exposes its tools automatically, so your agent can read the transcript and edits and use the same editing controls. Manage permissions through your assistant or browser. You can also edit manually.

For a smoother browser experience, we recommend clips of 10 minutes or less. Longer recordings may work, but the practical limit depends on your computer, available memory and browser storage, video resolution, and speech model. For hour-long recordings, split the file into shorter clips first. Keep this tab open while transcribing and exporting.

Detection can miss fillers or mistake speech. Listen before applying cuts. Use Undo or restore a removed section to reverse a cut.

Edits autosave in this browser. Download a project backup and keep the original recording. Reopen the backup and reconnect that recording to continue. Clearing browser data removes local projects.