Local Whisper
Local speech-to-text using OpenAI's Whisper model. Fully offline after initial model download. Supports multiple model sizes from tiny (39M) to large-v3 (1.5GB) with options for timestamps, JSON output, and language auto-detection.
🚀 Convert audio to text instantly with Local Whisper, OpenAI's powerful speech-to-text engine. Works completely offline after downloading the model once—no internet required, no data sent to servers. Choose from five model sizes to match your needs, from lightning-fast tiny (39MB) to ultra-accurate large-v3 (1.5GB).
💡 Perfect for transcribing meetings, interviews, voice notes, and podcasts. Add timestamps for precise word-level timing, or export as JSON for seamless integration with other tools. Supports auto-language detection across 99+ languages.
✨ Enjoy complete privacy and zero latency—your audio never leaves your device. Fast processing with flexible model options means you control the speed-vs-accuracy tradeoff.
Requirements
OpenAI's Whisper speech recognition model
PyTorch deep learning framework
Python CLI framework