TypeWhisper/typewhisper-mac
Local speech-to-text for macOS on-device AI, fully private, optional cloud
What it solves
TypeWhisper for Mac is a system-wide speech-to-text and AI text processing tool. It allows users to dictate text into any application, transcribe audio files, and use AI to transform that text through customizable workflows, all while providing a choice between local, on-device processing for privacy and cloud APIs for speed.
How it works
The application integrates multiple transcription engines (such as WhisperKit, Parakeet, and various cloud APIs) and LLM providers (including local Gemma 4 via MLX and cloud options like Groq and OpenAI). It uses global hotkeys for push-to-talk dictation and an extensible plugin system to add new engines or post-processors. It also features a local REST API and CLI tool for automation.
Who it’s for
It is designed for macOS users who need high-quality dictation and transcription, especially those who prefer local AI models for privacy or those who want to automate text transformation workflows based on the app or website they are using.
Highlights
- Hybrid Processing: Support for 11 different transcription engines, including local on-device models and cloud APIs.
- AI Workflows: Reusable transformations for translation, rewriting, and formatting that can be triggered automatically by app or website context.
- Extensibility: A plugin system for custom LLM, transcription, and TTS providers, plus a local HTTP API for external scripting.
- Personalization: A built-in dictionary for recognition accuracy, auto-learned corrections, and text snippets with placeholders.
- File Transcription: Batch processing of audio/video files with export options for SRT and WebVTT subtitles.