Freestyle offers a free open-source voice dictation pipeline
Freestyle is a app where you hold a hotkey, speak, release it, and the text is inserted at the cursor. The project first focused on letting people choose local s and add an optional cleanup step afterward. Local model support remains, but Freestyle Transcribe now gives users a default setup that works without choosing the best model mix themselves.
The current pipeline uses Whisper Large Turbo V3 to turn speech into text, then uses Qwen3-32B to clean up the result. It runs on Groq and Cloudflare. Reported delay is about 600 to 800 milliseconds.
The project claims this is close to Wispr Flow’s speed and more consistently accurate. Freestyle is free and open source, so it avoids Wispr Flow’s $12 per month cost.
Key points
- Freestyle turns spoken words into text and pastes them where the cursor is.
- It still supports , but also adds a ready-to-use Freestyle Transcribe setup.
- The pipeline combines Whisper Large Turbo V3 with Qwen3-32B for cleanup.
- It is hosted on Groq and Cloudflare with reported delay around 600 to 800 milliseconds.
- It is free and open source, unlike Wispr Flow’s $12 monthly plan.