Chirp-STT: Local Windows Dictation App Using ParakeetV3 Speech Recognition
By
whamp
Crisp on the outside, thoughtful on the inside. A keeper.
Summary
Chirp-STT is a Windows dictation application that runs fully locally using the ParakeetV3 speech-to-text model. The app is managed with uv and designed to work without requiring executable file permissions - if Python can run on the machine, Chirp can run. The article explains that ParakeetV3 was chosen because it offers accuracy comparable to Whisper-large-V3 (4.91 vs 5.05 multilingual WER) but is 17x faster and only requires a CPU, unlike Whisper models that need GPUs for comparable performance. The project is hosted on GitHub as an open-source development project.
Key quotes
· 3 pulledChirp is a Windows dictation app that runs fully locally using ParakeetV3 STT and is managed end-to-end with uv.
Chirp does not require the ability to run executable files (like .exe) on Windows. It was designed so that if you're allowed to run Python on your machine, you can run Chirp.
ParakeetV3 has indistinguishable accuracy from Whisper-large-V3 (multilingual WER 4.91 vs 5.05) but is 17x faster and only requires a CPU while Whisper models of comparable accuracy require GPU's.
You might also wanna read
Ringg launches Parrot: A speech-to-text model optimized for noisy, Hindi-heavy voice agent conversations
Ringg introduces Parrot, a speech-to-text model specifically designed for production-grade voice agents. Unlike standard STT models that per
Stet: Open-source macOS voice dictation app with local processing and AI refinement
A minimalist, open-source voice input app for macOS that processes speech locally, uses AI to refine dictation, and preserves the user's nat
Whispering: An Open-Source, Local-First Transcription App for Privacy-Conscious Users
Whispering is an open-source, local-first transcription app that prioritizes privacy by keeping audio data on-device. It supports both local
Ghost Pepper: Local Private AI for Text-to-Speech and Meeting Transcription on macOS
Ghost Pepper is a 100% local, private AI application for macOS that provides on-device text-to-speech and meeting transcription capabilities
Spoke: Private Voice-to-Text App for macOS with Local Speech Processing
Spoke is a macOS application that provides private voice-to-text transcription by running a local speech model on the user's device, ensurin
VibeSonic: On-Device AI Voice Dictation Software for Mac with Privacy Focus
VibeSonic is an AI voice dictation software for Mac that runs entirely on-device without cloud dependency or subscriptions. It uses on-devic
