TypeWhisper
speech-to-text with multiple engines, profiles
Last verified:
What is TypeWhisper?
TypeWhisper is a local-first, open-source speech-to-text application that runs on macOS, Windows, and iOS, designed to provide system-wide dictation and transcription without sending audio to the cloud. The app works with a hotkey to capture speech and paste transcribed text directly into any active text field, and it also supports file transcription, SRT export, and real-time streaming previews for selected engines. Key features include multiple selectable local engines (WhisperKit, Parakeet, Qwen3, Apple Speech and others as plugins), smart profiles for context-aware behavior, a local HTTP API for integrations, an editable dictionary and snippets, and a searchable history of past transcriptions. TypeWhisper targets privacy-conscious users, developers (for code dictation and comments), teams that need on-device transcription for sensitive data, and individuals who want a free, offline alternative to cloud dictation services.
TypeWhisper pricing
Pricing model: Freemium
TypeWhisper’s core app is free and open-source under GPLv3 for individual use, requiring no account or subscription; downloads for macOS and Windows are available with no monthly fees. The website lists Team and Enterprise commercial offerings for proprietary or managed use: a Team plan at 19 € / month (covers up to 10 devices) and an Enterprise plan at 99 € / month (unlimited devices) with device management, priority support, and licensing for proprietary use. Free personal use includes local engines and standard features, while paid plans add priority support, device management, proprietary licensing options, and assistance for company-wide rollouts.
TypeWhisper pros
- Runs completely on-device so audio never leaves your machine
- Open-source under GPLv3 for auditability and modification
- System-wide dictation via a global hotkey into any app
- Supports multiple local speech engines and engine plugins
- Real-time streaming preview with supported engines
- File transcription with SRT export for video workflows
- Local HTTP API enables automation and integrations
- Context-aware profiles to change behavior by app or task
- Searchable history and transcript management built-in
- Custom dictionary and snippets for improved accuracy
- No account, API keys, or subscriptions required for core app
- Options for LLM post-processing of transcripts
- Cross-platform availability (macOS, Windows, iOS alpha)
- Low ongoing cost — free to use for individuals
- Team and Enterprise plans available for managed deployments
TypeWhisper cons
- Some advanced engines require significant local compute (GPU/Apple NPU)
- Windows version is labeled beta (less mature than macOS)
- iOS is in early alpha with limited feature parity
- Setup for third-party engines can be technical for non-experts
- Accuracy depends on chosen local engine and hardware
- No cloud backup by default (user must handle syncing/history externally)
- Certain enterprise features require paid Team/Enterprise plan
- MacOS Apple-specific optimizations rely on Apple Silicon features
Frequently asked questions about TypeWhisper
Does TypeWhisper send my audio to the cloud?
No; TypeWhisper is designed to run speech recognition entirely on your device so your audio and transcriptions do not leave your machine.
Which platforms does TypeWhisper support?
TypeWhisper supports macOS (stable release), Windows (beta), and iOS (alpha), with the most mature feature set currently on macOS.
Do I need an account to use TypeWhisper?
No account is required for the free, open-source version — you can download and use the app locally without creating an account or providing API keys.
What speech engines can I use with TypeWhisper?
TypeWhisper supports multiple local engines and plugins such as WhisperKit (optimized for Apple hardware), Parakeet, Qwen3, Granite, and the built-in Apple Speech engine, with engine availability depending on platform and local hardware.
Can I transcribe audio files or export subtitles?
Yes; TypeWhisper can transcribe audio files and offers SRT export so you can generate subtitles from local transcriptions.
How do profiles and snippets work?
Profiles let you define context-aware settings (for example different engines, punctuation, or formatting by app), while snippets and a custom dictionary let you expand abbreviations and improve recognition of domain-specific terms.
Is there an API for automation or integrations?
Yes; TypeWhisper exposes a local HTTP API that allows automation, scripting, and integration with other local tools and workflows.
What are the Team and Enterprise plans for?
Team and Enterprise plans provide proprietary licensing, device management, priority support, and onboarding assistance intended for organizations that need audited deployments and company-wide rollouts.
How accurate is the transcription?
Accuracy varies by engine, hardware, microphone quality, and language; WhisperKit and other high-quality local engines generally deliver strong results but may require more powerful hardware for best performance.
Can I use TypeWhisper for coding or developer workflows?
Yes; TypeWhisper includes a Dictate Code use case with formatting prompts and snippets to help developers dictate comments, documentation, and simple code snippets directly into their editor.