Rescript vs Descript: a free, open-source alternative

Rescript vs Descript

Rescript is a free, open-source, offline alternative to the part of Descript most people use: cutting video and audio by editing a transcript. It transcribes locally, cuts as you delete words, strips fillers and silences, and can respeak a rewritten line in the original speaker's cloned voice — with no account and nothing uploaded. Descript is the larger product: it also records, collaborates, repairs audio, and edits multitrack. If you need a production suite, stay with Descript. If you need transcript-based cutting without a subscription, this covers it.

Try it in your browser Download desktop app Source

The Rescript editor: a transcript on the left with a struck-through word, video preview on the right, and a waveform timeline underneath.

The same workflow Descript is known for: the transcript on the left is the timeline, and deleting a word cuts the footage under it. The difference is that all of this is running on your machine.

Side by side

Rescript compared with Descript.

Feature comparison between Rescript and Descript
Feature	Rescript	Descript
Price	Free and open source for noncommercial use	Subscription; the free tier is capped
Where your media goes	Nowhere. It never leaves your device	Uploaded to Descript's cloud to be processed
Account required	None — open it and start editing	Required, even on the free tier
Watermark	Never	Removed on paid plans
Usage limits	No minute caps, no export quota	Transcription hours metered by plan
Works offline	Yes, once the model has downloaded a first time	Cloud-dependent for transcription and rendering
Source code	Public on GitHub — auditable	Proprietary
Delete words to cut footage	Word-level, with a live preview of the cut	Yes — this is the feature it invented
Filler word removal	One click across the whole file	Yes, plus "shorten word gaps"
Silence removal	One click, pauses of 0.3s and over	Yes
Speaker detection	Local diarization, speakers renameable	Yes
Rewrite a line and have it spoken	Yes — cloned from that speaker's own audio, on-device	Overdub — paid, and the voice is trained and stored in their cloud
Noise removal / audio repair	No — nothing like Studio Sound	Studio Sound, background noise removal
Records video or audio	No — bring a file you already have	Screen and camera recording built in
Multitrack, layers, effects	No — one clip, transcript-first	Multitrack, layers, transitions, titles
Team collaboration	No — projects are local to your machine	Real-time collaboration, comments, sharing
Burned-in captions	Exports SRT/VTT to burn in elsewhere; no styled overlay yet	Styled, animated, burned-in captions

Details about other products come from their own public pricing and documentation pages and change often — check theirs before you decide. Corrections are welcome as a GitHub issue.

Same workflow, different trust model

Both put a transcript where the timeline usually goes: select a sentence, press delete, and the footage goes with it. Rescript has the same one-click cleanups — strip every "um" in the file, cut every pause over 0.3 seconds.

What differs is where your media goes. Descript uploads it, which is exactly what makes its collaboration and cross-device features possible. Rescript uploads nothing: Whisper runs in your browser through WebGPU, diarization runs locally, and the render is ffmpeg on your own CPU. After the models download once, you can pull the network cable.

Overdub vs Regenerate

Descript's Overdub fixes a flubbed line in a voice trained on your recordings. It's paid, and the voice model is trained and stored on their servers.

Rescript's Regenerate does the same job on-device, cloning from about ten seconds of that speaker's own clean audio already in the project. No training session, nothing uploaded, and it works for any speaker in a multi-person recording — not just you. The new line is fitted to the exact span it replaces, so nothing downstream shifts.

Which one

They're good at different jobs.

Choose Descript if…

You record inside the tool — screen, camera, or remote guests.
You need multitrack production: layers, transitions, titles, effects.
A team edits the same project and comments on it.
Studio Sound is rescuing room audio you can't re-record.

Choose Rescript if…

You want the transcript workflow without a subscription.
The footage is confidential and can't be uploaded — interviews, legal recordings, unreleased work.
You edit long files and keep hitting a transcription-hours cap.
You'd rather read the code than trust a policy page.

FAQ

Questions people actually ask

Is Rescript really a free Descript alternative, or is there a catch?

It's free for noncommercial use under the PolyForm Noncommercial 1.0.0 license, with no feature gates, no trial, no transcription cap, and no watermark. Everything that runs on your own device stays free. Commercial use requires a paid license.

Can Rescript import a project from Descript?

Not a Descript project file, but you can export captions from Descript as SRT or VTT and import them into Rescript alongside the original media. Rescript will edit against your transcript instead of running Whisper.

Does Rescript have anything like Studio Sound?

No. There's no noise removal, de-reverb, or audio enhancement. Rescript cuts and re-encodes what's already in the file. If your source audio needs repair, do that first in another tool.

Is Rescript's transcription as accurate as Descript's?

It uses Whisper Base or Whisper Small, which handle clear speech well but won't match a tuned cloud pipeline on difficult audio. Both the words and their timings are editable, so you can correct anything it gets wrong and the cuts follow your correction.

Keep comparing