typewhisper-mac
A macOS speech-to-text app with local models, optional cloud APIs, system-wide dictation and on-device AI text workflows.
Alternative to:
TypeWhisper is a macOS dictation and transcription app for people who want to turn speech into text across their apps. Local speech models keep audio on your Mac; optional cloud transcription sends it to the selected provider.
- Dictation across apps: Insert speech into the active app, preview partial transcripts with WhisperKit, and undo the last dictation or restore its raw text.
- Recordings and subtitles: Batch-transcribe audio and video files, including files selected in Finder, and export timestamped SRT or WebVTT subtitles.
- Text workflows: Apply reusable rewriting, extraction, translation and formatting workflows to dictation, selected text or clipboard contents, with triggers for particular apps and websites.
- Transcript corrections: Search and edit transcription history, apply dictionary corrections automatically, and import vocabulary from Wispr Flow, Handy or compatible CSV files.
- Plugin integrations: Connect workflows to Obsidian, Linear and MCP tools, or extend the app with compatible transcription engines and AI providers.
The app is open source under GPL-3.0. Local speech engines include WhisperKit and Parakeet, with compatible model imports available through local plugins. Apple Translate handles translation on-device. The Local LLM plugin uses MLX on Apple Silicon and recommends Gemma 4 E2B/E4B models. Apple Intelligence requires macOS 26 or later. Optional cloud providers include Groq, OpenAI and xAI/Grok; workflows using cloud LLMs send their text to those services.