See It In Action

Your command center

The main screen puts everything within reach. Switch microphones, swap between six AI models of varying speed and accuracy, and change languages — all without opening a menu. A quick-access hotwords button lets you add domain-specific terms on the fly.

The last transcription is always visible at the bottom, alongside a sample counter showing how many voice recordings you have stored for personal model training. From here you can open Settings, view History, correct the last transcription, or start recording with a single click.

Closing the window hides the app to the system tray — it keeps running in the background, ready the moment you press your hotkey.

Press, speak, done

Press your hotkey and the app starts listening. A live timer shows recording progress and remaining time. When you release (or press again in toggle mode), the audio is transcribed and typed directly into whatever application has focus.

For longer dictation, streaming mode transcribes and types text in real-time as you speak — no waiting for the full recording to finish.

Teach it your vocabulary

The Hotwords dialog lets you feed the AI a list of words it should prioritize — names, technical terms, brand names, or any uncommon vocabulary. Just type a word and press Enter.

The model uses these hints during transcription to avoid misrecognizing important terms. No retraining required — hotwords take effect immediately.

Fix mistakes, improve the model

The Correct Transcription dialog appears when you tap the Correct button on the main screen. It shows the original transcription alongside an editable correction field.

When you save, the corrected text and its matching audio clip are automatically added to your training data — building a personalized dataset that improves the model's accuracy for your voice over time.

Full control over input

The Audio tab is where you configure how you interact with the app. Set your recording hotkey from a list of shortcuts, choose between hold-to-record and toggle-record modes, or assign a mouse side button for hands-free control.

A volume slider adjusts input gain, and the text injection backend can be changed if certain applications don't accept the default typing method. At the bottom, a toggle enables the built-in AI noise reduction powered by DeepFilter, with an adjustable attenuation slider that controls how aggressively background noise is suppressed.

36 languages, one click

The Language tab sets which language the AI listens for. Choose from 36 supported languages or leave it on auto-detect.

You can also build a list of favorite languages that appear in a quick-switch dropdown on the main screen — perfect if you regularly dictate in more than one language.

Your voice, your dataset

The Training Data window lists every voice sample you have recorded. You can play back any sample, read its transcription, and delete entries you no longer want.

These samples form the dataset used when you fine-tune the model to your voice. The more you correct and record, the better the model gets at understanding you.

Personalize the AI

The Training tab is where you will fine-tune the AI model on your own voice. Local LoRA training adapts the model to your accent, vocabulary, and speaking style — all without sending data anywhere.

This feature is under active development and will be available soon.

Everything you said, searchable

The History tab controls how transcriptions are stored. Enable session history to keep a log that clears when the app closes, or switch to persistent history to save transcriptions across sessions.

The Transcription History window displays a timestamped log of everything the app has transcribed. A Clear History button lets you wipe the log at any time. All history is stored locally on your device.

Make it yours

The Appearance tab lets you choose between a Dark and Light theme and change the interface language. Theme changes take effect after a restart.

Your data stays yours

The Privacy tab gives you control over your data. Enable local encryption to protect your stored training samples. A privacy notice confirms that all audio is processed entirely on your device and is never sent to any server.

Stay up to date

The Updates tab shows your current version, links to the website, and provides contact details for support. A Report a Bug button lets you send feedback directly to the development team. Updates are checked automatically on startup.