Latest Version: 1.0.0
Speech-to-text inside REAPER, now without Docker! ReaSpeech 1.0 uses ReaSpeech Lib for transcription and installs through ReaPack. Recognition runs in the background with progress reporting and cancellation.
The release’s ReaSpeech.lua asset is the bundled script for manual installation through REAPER’s Actions window. ReaImGui 0.10 or newer and a ReaSpeech Lib backend are still required. ReaPack is the recommended way to install and receive updates.
Upgrading from 0.7.x and other Docker-based versions
-
Close the old ReaSpeech window. Add the Team Audio repository to ReaPack:
https://github.com/TeamAudio/reascripts/raw/main/index.xml -
Install ReaSpeech, one ReaSpeech Lib backend, and ReaImGui 0.10 or newer. Restart REAPER to load the extension.
-
Run the ReaPack-installed ReaSpeech script from REAPER’s Actions window. Update any keyboard shortcuts and toolbar buttons to use this action. Your old downloaded script is a separate copy and still expects the Docker web service; installing through ReaPack does not update that copy.
-
Select a media item and click Transcribe Selected Items to check the new setup. Once it works, stop your old ReaSpeech Docker container and disable any automatic startup for it. You can then remove the old script’s action after updating your shortcuts.
Docker and the local web server are no longer required for ReaSpeech. There is no need to uninstall Docker or remove containers used by other tools.
Need the legacy version?
The older images remain available on Docker Hub, with setup instructions in the 0.7.3 documentation. We will continue hosting these images for a limited time, so please plan to move to the new version.
Ready to change the way you work with voice?
Meet ReaSpeech - a Whisper-powered ReaScript that brings speech transcription and navigation directly to REAPER. Transcription runs on your own machine, with no Docker container or Python backend to set up.
Installation
- In REAPER, open Extensions → ReaPack → Import repositories… and paste https://github.com/TeamAudio/reascripts/raw/main/index.xml Don't have ReaPack yet? Get it here.
- Install ReaSpeech, exactly one ReaSpeech Lib backend from the table below, and ReaImGui 0.10 or newer, then restart REAPER.
- Run ReaSpeech from the Actions window, select one or more media items, and click Transcribe Selected Items. Models download on first use.
| Backend | For |
|---|---|
| ReaSpeech Lib (CPU) | Windows and Linux (x86_64), no GPU required |
| ReaSpeech Lib (Metal) | macOS on Apple silicon (M1 and later) |
| ReaSpeech Lib (CUDA 13) | Windows and Linux with a current NVIDIA driver |
| ReaSpeech Lib (CUDA 12) | Windows and Linux with an older NVIDIA driver |
Intel Macs aren't supported at the moment.
Coming from the Docker version? Follow our migration instructions.
ReaSpeech is open source!
Why just be a ReaSpeech super fan when you can be a ReaSpeech super contributor? Visit the GitHub Project for the source code and documentation of our development and contribution processes.
Thoughts? Questions? Suggestions? Lamentations on the state of audio plugin development and usage? Come commiserate with us on Discord.