Dictato turns speech into text on your Mac using on-device machine learning models. The app lives in your menu bar and is always ready - you configure a global hotkey, press and hold it to speak, then release to have your speech transcribed and inserted wherever your cursor is positioned.

Dictato offers three transcription engines: Parakeet (NVIDIA's open-source ASR model optimized for speed and accuracy), Whisper (OpenAI's speech recognition model with broad multilingual coverage), and Apple SpeechAnalyzer (Apple's built-in engine). The app supports 25-99 languages depending on the engine chosen, features automatic language detection, and includes optional proofreading and translation capabilities.

All processing happens locally on your Mac's hardware, leveraging Apple Silicon's Neural Engine for fast inference with latency as low as 80 milliseconds. The audio from your microphone is processed in real time and discarded immediately after transcription, requiring no internet connection or cloud servers.

Dictato enables users to write 3x faster by speaking at 150 WPM instead of typing at 40 WPM. It eliminates Mac's built-in 60-second timeout limitation and works universally across macOS applications including Gmail, Slack, VS Code, Notion, Pages, and anywhere you can type.

The app is designed for writers, developers, and professionals who need fast, private dictation. It's particularly suitable for sensitive use cases like legal documentation, medical notes, and confidential work since no audio or transcription data ever leaves the device.