VoicePhoto

VoicePhoto is a small Android app that captures photos or records video using voice triggers. This is a university project and uses Vosk on-device speech recognition models. Configure one or more custom keywords (English or Czech) and the app will take a picture or start/stop recording when it detects them.

Key features

• Trigger photo capture or video recording with voice keywords
• Support for English and Czech (on-device Vosk models included)
• Uses AndroidX CameraX for camera lifecycle and capture
• Simple settings to add/remove keywords and choose language
• Privacy-focused: no account required; processing can run on-device

How it works (high level)

1. App unpacks a Vosk speech model from assets and creates a recognizer.
2. Microphone audio is fed into the recognizer; partial results are monitored for configured keywords.
3. When a keyword is detected the app triggers CameraX to take a photo or toggle video recording.
4. Captures are saved to device storage; optional beeps/flash give user feedback.

Quick start — build & run

Prerequisites:

• Android Studio (Arctic Fox or later recommended)
• Java / Android SDK matching project config (check app/build.gradle)
• Device or emulator with camera and microphone

To build and run:

1. Open the project in Android Studio.
2. Let Gradle sync and download dependencies.
3. Run the app on a device (recommended) or an emulator with camera/mic support.

Permissions

The app requires:

• RECORD_AUDIO — for keyword detection
• CAMERA — for photo/video capture
• WRITE_EXTERNAL_STORAGE / READ_EXTERNAL_STORAGE — to save and access photos (if applicable)

Grant these permissions on install or at runtime.

Configuration & usage

• Open Settings to add keywords and choose language (English / Czech).
• Use short, distinct words/phrases to reduce false positives.
• Use in quiet environments for best results.
• Toggle flash or sound feedback in the UI.

Troubleshooting

• No recognition: verify microphone permission and model loaded successfully (logcat shows model/init messages).
• False triggers: shorten or change keywords, test in quieter environment.
• Camera errors: ensure CameraX-compatible device or emulator and proper permissions.

Project structure (important files)

• app/src/main/java/vut/example/voskapp/MainActivity.java — main app logic, recognition and capture
• app/src/main/java/vut/example/voskapp/SettingsActivity.java — settings and keywords
• app/src/main/java/vut/example/voskapp/HelpActivity.java — help UI
• models/ — Vosk model assets (English/Czech) used by recognizer
• app/build.gradle — app build configuration

Contributing

Bugs, improvements and pull requests are welcome. Open an issue describing the problem or desired change and include logs or reproduction steps when possible.

License & credits

Vosk speech recognition (models) and CameraX are used under their respective licenses — check models/ and build files for specifics.