Privacy Policy

Effective date: March 28, 2026

This privacy policy describes how the Speech-to-Text Android application ("App") handles your information. We are committed to protecting your privacy and being transparent about our practices.

Key point: All speech recognition happens entirely on your device. Your audio and transcripts are never sent to our servers.

1. Information We Do Not Collect

The App performs all speech-to-text processing locally on your device using offline language models. We do not collect, store, or transmit:

Audio recordings or microphone input
Transcribed text or transcript history
Personal information or identifiers
Contacts, location, or browsing data
Keyboard input when using the SpeechToText keyboard

2. On-Device Data

The following data is created and stored only on your device:

Transcripts: All transcribed text is saved locally in the app's private database and is never transmitted off the device.
Audio recordings: If you choose to save recordings, they are stored locally on your device.
App preferences: Settings such as your chosen language model and display preferences are stored locally.

You can delete your transcripts and recordings at any time from within the App.

3. Network Access

The App requires network access only to download speech recognition language models from the Vosk project servers. No personal data, audio, or transcribed content is sent during these downloads. Once a model is downloaded, the App works fully offline.

4. Crash Reporting (Optional)

If configured, the App may use Sentry for anonymous crash and error reporting. This data includes:

Device model and OS version
App version and crash stack traces
General performance metrics

Crash reports do not include any audio, transcripts, or personally identifiable information. This feature is optional and serves solely to help us improve App stability.

5. Permissions

The App requests the following permissions:

Microphone: Required for live speech-to-text transcription. Audio is processed entirely on-device.
Internet: Used solely to download language models.
Notifications: Used to inform you about model download progress.

6. Third-Party Services

The App uses the following third-party components:

Vosk: Open-source speech recognition engine that runs locally on-device. Vosk servers are contacted only to download language model files.
Sentry (optional): Anonymous crash reporting service. No personal data is shared.

7. Children's Privacy

The App does not knowingly collect any personal information from children under 13. Since the App does not collect personal data from any user, it is safe for use by all ages.

8. Data Security

Since your data never leaves your device, it is protected by your device's built-in security measures, including device encryption and app sandboxing provided by Android.

9. Changes to This Policy

We may update this privacy policy from time to time. Any changes will be reflected by updating the "Effective date" at the top of this page. We encourage you to review this policy periodically.

10. Contact Us

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

shezadamanulla@gmail.com