Your stream Live
early development

Local subtitles for your own broadcasts.

Caption any audio source, integrate directly with OBS, customized for your stream. You are in full control. Free, Fully local, open-source, no cloud.

Still in early development, so expect lots of changes.

Download for macOS All releases
v0.2.16AGPL-3.0macOS · Windows · Linux GitHub
and this is

Creator First

There are a handful of good general subtitling programs out there, but none of them are built with creators and streaming in mind.

Free and Opensource

This means no subscriptions, no cloud latency, no API keys.

Privacy Focused

Data never leaves your computer. All audio and text processing happens where you have full control over it.

What it does

Scrybe runs local Speech-To-Text models on your machine to create live captions that you can add to OBS as a browser source.

All processing happens on your machine. Your audio never leaves your computer. This means no API keys, no cloud latency, no subscriptions.

audio in whisper.cpp text out localhost only
[00:02.6] the captions you are reading right now are generated on this machine
[00:05.2] no api keys, no cloud latency, no subscriptions
[00:07.8] swap the model mid-stream if your gpu can take it
[00:10.4] alright, back to the game
Planned

Cleaner live transcripts

More natural caption segments with silence and speaker-aware breaks, steadier updates, accurate timestamps, and speaker identification.

Setup

Four steps from install to on-stream subtitles.

Scrybe is a standalone desktop app, install it like any normal program. Choose from a set of model presets (or bring your own) and start transcribing.

01

Install and launch Scrybe.

Scrybe is a standalone desktop app, install it like any normal program.

02

Customize the subtitle box's positioning, color, and size.

Adjust text alignment, color, font size, transparency, etc and the overlay will update automatically.

03

Add the overlay URL to OBS

localhost:3030/app/v1/overlay
04

Press Start and captions appear.

Go live

Under the hood

Powered by whisper.cpp.

Scrybe embeds it directly. Audio goes in, text comes out, nothing is sent remotely.

whisper.cpp is Georgi Gerganov's port of OpenAI's Whisper speech-recognition model to plain C/C++. More at github.com/ggml-org/whisper.cpp.

No runtime, No PythonRust/C++
Metal (Apple), CUDA (Nvidia), Vulkan (cross-platform)CPU + GPU
No API bills, No account creation$0

Download for macOS, Windows, Linux · Debian/Ubuntu, or Linux · Fedora/RHEL.