Hold a hotkey, speak naturally, and let AI transcribe & polish your words — then auto-type the result into any app.
Native macOS experience, from speech to polished text
Hold Option, speak, release to send. Powered by Aliyun FunASR with accurate Chinese & English transcription.
Removes filler words, fixes grammar, turns casual speech into clean text. Powered by Qwen.
Polished text is typed directly into the focused app — browsers, editors, chat apps, anything.
Glassmorphism overlay with live waveform and real-time transcript preview. Always on top.
Bring your own Aliyun API key. Credentials stay in macOS Keychain — we never see them.
Silero VAD runs entirely on-device via ONNX Runtime. No network needed for voice detection.
From speech to polished text, in seconds
Press and hold the Option key (or your custom hotkey) to start recording.
Just talk. AI handles filler words, grammar, and formatting automatically.
Let go. Polished text appears instantly in whatever app you're using.
Your voice goes directly to Aliyun, never through us.
We run no backend. Audio goes directly from your Mac to Aliyun for transcription.
API keys are stored in macOS Keychain, encrypted by the system. We have no access.
No accounts, no analytics, no telemetry. We collect literally nothing.
MIT Licensed. Every line of code is on GitHub — audit it yourself.
Free, open source, no servers. Just download and start talking.