EasyWhisperUI

Transcribe audio and video on your own hardware with Whisper and GPU acceleration. Works offline after downloading the required tools and models.

What should I work on next? Vote for the features you want most.

Download

Windows uses Vulkan on supported GPUs; macOS uses Metal. CPU-only processing is also available.

Release notes

A note from me

EasyWhisperUI has come a long way—from the original C++ app to a full rewrite in Electron, and now a redesigned interface with more reliable setup and GPU checks.

I've heard your feedback, issues, and the many feature requests. Countless hours have gone into rebuilding, testing, and making the app easier to use. Your bug reports and suggestions help shape what I work on next. There's more to come!

Special thanks to past contributors, such as celerycoloured who authored the initial macOS release back when it was C++, and to everyone who has supported the project through donations, bug reports, and feature requests. Your contributions helped make EasyWhisperUI better.

Features

• Private, local transcription — turn audio and video into text on your own computer.
• GPU acceleration — Vulkan on Windows and Metal on Apple Silicon, with automatic verification and a CPU-only option.
• Languages and subtitles — multilingual transcription, translation into English, and TXT/SRT exports.
• Batch processing — drag in multiple files, skip individual items, or stop the queue.
• Live transcription — transcribe microphone audio in real time (beta).
• Model choice — download, manage, or load custom models, with color-coded memory estimates.
• Clear download progress — see size, speed and time remaining, and cancel downloads when needed.
• Light, Dark and Auto themes — translucent backgrounds and a spacious output console.
• Faster retries — cached audio avoids repeated conversion when retrying or changing models.
• Simple setup and repair — guided installation and clean reinstall on Windows/Linux, preserving models and settings.
• Advanced controls — extra Whisper arguments, built-in CLI help, and accessible setup logs.

Quick start

1. Run the Windows installer, or open the Mac DMG and drag EasyWhisperUI into Applications.
2. Launch the app and let initial setup finish. macOS includes Whisper and FFmpeg.
3. Choose your model, input language, and output formats. Use a multilingual model for non-English audio; .en models are English-only. Missing models download automatically. For a local model, choose custom → Select Model File.
4. Click Open or drop files into the window. Transcription starts automatically; exports are saved beside the original files.

To translate into English, choose a multilingual model such as medium, select the source language, and add --translate in Arguments.

Using your transcript

Give the TXT file to your favorite LLM for summaries, study notes, or a cleaned-up transcript. Export SRT to add subtitles in a video editor or player.

Troubleshooting

On Windows/Linux, try Settings → Clean reinstall. On Mac, restart the app. If the problem continues, use Settings → Show log file and open an issue with the error, log, OS, CPU, GPU, RAM, app version, and model. If no setup log exists, copy the console error.

The Mac app is ad-hoc signed and not notarized. If macOS blocks it, try System Settings → Privacy & Security → Open Anyway. For a “damaged” message, follow the release's first-opening instructions.

Development

Requires Node.js 22.12+; Node.js 24 is recommended. Mac builds also require Homebrew and Xcode Command Line Tools. Clone with submodules, then:

cd electron
npm install
npm run dev

Use npm run dist to package the app into build/electron-dist at the repository root. See the developer README for architecture and platform setup.

Support

I'm a solo developer, and countless hours went into refining EasyWhisperUI. Donations are the project's only source of funding. If you find it useful, please consider supporting continued development.

Thank you to my supporters:

• Craig H — $50
• Eric De Vet — $10
• Jan P — $5
• Minh P — $5
• Rödvarg R — $2

Credits

whisper.cpp by Georgi Gerganov · FFmpeg · Windows FFmpeg builds · macOS FFmpeg builds · SDL2 · electron-builder

License

Copyright (c) 2026 Mehtab Mahir
All rights reserved.

This software is proprietary and the following is not allowed for commercial purposes:
it may not be copied, modified, distributed, or used without explicit permission from the author.

Those actions are permitted for personal use ONLY.

This application includes the following open-source components:

---

whisper.cpp by Georgi Gerganov
License: MIT
[https://github.com/ggerganov/whisper.cpp](https://github.com/ggerganov/whisper.cpp)

---

FFmpeg
License: depends on the build; the bundled macOS binary is GPL 3.0 or later.
Windows setup downloads a GPL build from BtbN.
[https://ffmpeg.org](https://ffmpeg.org)
Windows builds: https://github.com/BtbN/FFmpeg-Builds
macOS builds: https://ffmpeg.martin-riedl.de/

The FFmpeg binary is provided as a separate file and may be replaced with a compatible version.

---

SDL2 (bundled on macOS)
License: zlib
https://www.libsdl.org/