What is TypeWhisper?
TypeWhisper is a speech to text app that runs entirely on your device. It transcribes your voice and pastes the result into whatever app you have open. No audio leaves your machine during transcription. It covers daily dictation, audio and video file transcription, and automated workflows. A global hotkey triggers recording from anywhere on your system. TypeWhisper is free and open-source, available on macOS, Windows, and iOS.
Features & Benefits
- System-Wide Dictation: Activate voice to text in any app using a global hotkey. Supports push-to-talk, toggle, and hybrid modes.
- On-Device Processing: Run the full speech to text engine locally. No telemetry, no data collection, no network requests during transcription.
- Multiple Speech Engines: Choose from WhisperKit, Parakeet TDT, Voxtral, Qwen3 ASR, IBM Granite Speech, Apple Speech, and ONNX-based engines. Add cloud engines via plugins.
- File Transcription: Drag and drop audio or video files for transcription. Export subtitles as SRT or WebVTT.
- Live Transcript: Display real-time transcription in a floating window during meetings or presentations.
- Per-App Profiles: Automatically switch language, engine, and post-processing rules based on the active app or website.
- Post-Processing Pipeline: Apply AI-powered text refinement using LLM providers after dictation.
- Dictionary and Snippets: Define custom terms, corrections, and text expansions with dynamic placeholders.
- Workflows: Reorder workflow steps with drag and drop. Run watch-folder jobs. Override engine or model per dictation session.
- History and Export: Search past transcriptions and export them as Markdown or JSON.
- Local HTTP API: Control the dictation app through scripts, shortcuts, and automation tools via a local API.
- Plugin System: Extend the speech to text app with 20+ cloud providers including OpenAI, Groq, Gemini, Deepgram, and Speechmatics.
- iOS Keyboard Extension: Use voice to text input in any iOS app through a custom keyboard.
- iOS Share Extension: Send audio or video files from other iOS apps directly to TypeWhisper for transcription.
What can TypeWhisper do?
- Transcribe speech to text without internet
- Dictate text into any app on Mac or Windows
- Transcribe audio files to text
- Transcribe video files to text
- Export transcriptions as SRT subtitles
- Export transcriptions as WebVTT subtitles
- Run speech recognition entirely on-device
- Switch dictation language per app
- Expand text snippets while dictating
- Search transcription history
- Automate transcription with watch folders
- Control voice dictation via local HTTP API
- Translate speech with built-in engine support
Real-World Applications
Writing in any app gets faster when a dictation app handles the typing. Drafting emails, filling out forms, or adding notes to a project tool can all happen by voice. TypeWhisper pastes the transcription directly into the active window, so the workflow stays intact.
Journalists, researchers, and content creators may use the file transcription feature to turn recorded interviews or field audio into searchable text. Dropping a video file and exporting an SRT subtitle file can cut hours from a post-production process.
Privacy-sensitive environments benefit from a speech to text app that keeps all processing local. Medical, legal, and financial professionals who dictate notes can do so without sending audio to a third-party server. Per-app profiles let the tool behave differently in each application without manual adjustment.
Developers and power users can connect TypeWhisper to their existing tools through the local HTTP API. Automation scripts can trigger recordings, retrieve history, or control dictionary behavior. The plugin system extends the voice to text app to cloud providers when higher accuracy or broader language support is needed.
