Buzz Captions Offline Audio Transcription and Translation Powered by Whisper

What is Buzz Captions?

Buzz Captions provides offline audio transcription and translation capabilities, leveraging the power of OpenAI's Whisper model. Users can import audio and video files and conveniently export the resulting transcripts into multiple formats, including CSV, SRT, TXT, and VTT. The tool also features live transcription and translation directly from a computer's microphone, although performance may vary based on system resources and selected model settings.

Available across Windows, Linux, and macOS (Intel), Buzz Captions supports several Whisper implementations like Whisper.cpp and Faster Whisper, along with Whisper-compatible Hugging Face models and the official OpenAI Whisper API. The macOS version boasts a native look and feel, incorporating features such as transcript search, audio playback synced with the transcript, and inline editing capabilities. As an open-source project, its code is accessible on GitHub.

Features

Offline Processing: Transcribe and translate audio/video without needing an internet connection (except for API usage).
Multi-Format Support: Import audio/video files and export transcripts to CSV, SRT, TXT, and VTT.
Live Transcription: Capture and transcribe audio in real-time from microphone input.
Extensive Language Support: Handles transcription and translation in over 90 languages.
Multiple Model Compatibility: Works with Whisper, Whisper.cpp, Faster Whisper, Hugging Face models, and the OpenAI Whisper API.
Cross-Platform Availability: Runs on Windows, Linux, and macOS (Intel).
macOS Enhancements: Native interface, transcript search, audio playback, and inline editing for Mac users.
Open Source: Code available on GitHub for transparency and community contribution.