EasyWhisperUI: A Cross-Platform Desktop App for Whisper Model Transcription

Summary

EasyWhisperUI is a fast, local desktop application designed for transcribing audio and video using the Whisper model. It offers GPU acceleration across Windows, macOS, and Linux, providing a user-friendly interface for various transcription tasks. The application supports features like live transcription, batch processing, and translation, making it a versatile tool for media processing.

Use at your own risk

OSRepos shares public repositories for knowledge and discovery only. Any installation, execution, configuration, or use of code from these repositories is the user's own responsibility. Always review the repository, source code, dependencies, licenses, and security implications before running or installing anything. OSRepos is not responsible for issues, damages, or losses resulting from third-party repositories.

Introduction

EasyWhisperUI is a fast, local desktop application designed to transcribe audio and video files using the powerful Whisper model. Optimized for various GPUs, it provides an easy-to-use interface across Windows, macOS, and Linux. The project recently underwent a significant migration to an Electron architecture, built with TypeScript, ensuring a consistent cross-platform user interface and a more robust development workflow. Key features include GPU acceleration, live transcription (beta), batch processing, and translation support for over 100 languages.

Installation

Getting started with EasyWhisperUI is straightforward. You can download the latest installers directly from the project's GitHub Releases page.

Windows

Download the latest Windows installer from Releases.
Run the installer, which will install the application per-user under app data.
Desktop and Start Menu shortcuts will be created.

macOS

Download the .dmg file from Releases.
Open the .dmg and drag EasyWhisperUI into your Applications folder.

Linux

Download one of the Linux artifacts (e.g., .AppImage, .deb, or .rpm) from Releases.
Install or run it using your distribution's standard workflow.
On first launch, EasyWhisperUI will install required dependencies and compile Whisper binaries. If any issues arise, please open an issue with logs on GitHub.

Examples

Using EasyWhisperUI for transcription is intuitive:

Basic Transcription: Simply drag and drop your audio or video files into the application, or use the "Open With" integration.
Batch Processing: Queue multiple files for sequential transcription.
Translation: Translate transcribed content into over 100 languages.
Model Selection: Choose from various Whisper models like tiny, medium-en, or large-v3 to suit your needs.
Custom Models: Select custom in the model selector, then choose your local whisper.cpp-compatible model file (.gguf, .ggml, or .bin) via the file picker.
Output Formats: Export transcriptions in .txt or .srt (with timestamps) formats.

Why Use EasyWhisperUI?

EasyWhisperUI stands out for several reasons:

Optimized Performance: It leverages GPU acceleration (Vulkan on Windows, Metal on macOS Apple Silicon) for significantly faster transcription, falling back to CPU if GPU is unavailable.
Cross-Platform Consistency: The Electron-based UI ensures a unified experience across Windows, macOS, and Linux.
Rich Feature Set: From live transcription and batch processing to extensive language translation and custom model support, it covers a wide range of use cases.
User-Friendly Interface: Designed for ease of use, it simplifies complex transcription tasks.
Reliable Foundation: The recent rewrite to Electron and TypeScript provides a stable and maintainable platform for future enhancements.