What is Whispering?
Whispering is an open-source, local-first transcription app. Use local and cloud models, chain custom transforms, and most importantly, keep your audio local on-device. Fast, ergonomic, and MIT-licensed. Let’s make closed-source apps obsolete. 🚀
Problem
Users rely on closed-source transcription apps that require uploading audio to the cloud, facing privacy risks and lack of control over their data. Existing solutions often depend on internet connectivity and offer limited customization of transcription models.
Solution
An open-source, local-first transcription app enabling users to process audio on-device with optional cloud models. Features include chaining custom transforms (e.g., noise reduction, speaker diarization) and ensuring audio data remains local. Example: Transcribe offline while applying privacy filters.
Customers
Developers, privacy-focused researchers, and journalists needing secure audio processing. Open-source advocates and enterprises handling sensitive data.
Unique Features
Hybrid local/cloud model support, MIT-licensed codebase, modular transform pipeline, and offline-first architecture. Unlike cloud-only tools, it allows granular control over data flow.
User Comments
Essential for confidential interviews
Runs smoothly on older hardware
Game-changer for medical transcription
Love the export plugins
Finally replaces proprietary dictation tools
Traction
4.7/5 stars on Product Hunt (322 votes), 1.2k GitHub stars, 48 active contributors. Used by 500+ healthcare clinics per founder interviews.
Market Size
The global speech-to-text market is projected to reach $21.5 billion by 2026 (MarketsandMarkets). Open-source AI transcription specifically grew 300% YoY per Gartner.
