Privacy Policy

Overview

Voxxy is designed with privacy at its core. Your voice data is processed on your device and never sent to external servers by Voxxy.

Data Collection

Voxxy does not collect, store, or transmit any personal data.

• No analytics or tracking
• No user accounts
• No cloud services
• No crash reporting sent externally
• No advertising

Speech Recognition

Voxxy uses OpenAI's Whisper, an open-source speech recognition model, to transcribe your voice. Whisper runs entirely on-device through WhisperKit, which uses Apple's Core ML framework for hardware-accelerated inference on your Mac.

All transcription happens locally on your Mac. Your voice is never sent to OpenAI, Apple, or any other external server. Voxxy does not depend on Siri, Dictation, or any system speech recognition service. It works even with those features disabled.

A small Whisper model is bundled with the app so Voxxy works immediately on first launch without any download. If you choose a larger, more accurate model in settings, Voxxy downloads it one time from the WhisperKit model repository on Hugging Face and caches it locally. This is the only network request Voxxy ever makes, and it only contains the public model name. No personal data is transmitted.

Microphone Access

Voxxy requires microphone access to record your voice for transcription. Audio is only recorded while you actively hold your configured hotkey or press the record button. No audio is stored permanently. It is discarded immediately after transcription.

Transcription History

Your transcription history is stored locally on your Mac using UserDefaults. It never leaves your device. You can clear your history at any time from within the app.

Accessibility Permission

Voxxy requests Accessibility permission to detect your configured hotkey (FN, Option, Command, Control, or Shift) globally and to paste transcribed text into active text fields. This permission is only used for keyboard event detection and simulating paste commands.

Third-Party Services

Voxxy does not use any third-party analytics, advertising, tracking, or data-processing services. Your voice, transcripts, usage patterns, and personal information are never sent to anyone.

There is exactly one situation in which Voxxy contacts an external server: when you choose a Whisper model that is not bundled with the app, Voxxy downloads the model files from Hugging Face, a public model repository. The request contains only the public model name (for example, openai_whisper-small.en) and standard HTTP headers. No personal data, no account identifiers, no voice data, and no transcripts are ever transmitted.

Downloaded models are cached locally and reused. If you only use the model bundled with the app, Voxxy never makes any network request at all and can be used fully offline. Hugging Face's own privacy practices are governed by their privacy policy.

Children's Privacy

Voxxy does not knowingly collect data from children under the age of 13. Since Voxxy collects no personal data at all, there is no risk of children's data being collected.

Changes to This Policy

We may update this privacy policy from time to time. Any changes will be reflected on this page with an updated "Last updated" date.

Contact

If you have questions about this privacy policy, please contact: