Voice Input

The mobile app supports hands-free voice interaction through a hold-to-talk interface. Press and hold the microphone button to record, release to send.

Hold-to-talk

1. Long-press the microphone icon in the chat input bar
2. A waveform animation appears showing audio levels
3. Speak your message naturally
4. Release your finger to send — the audio is transcribed and sent as a chat message
5. The agent responds with a streaming text reply

Tip

Swipe away from the microphone icon while holding to cancel the recording without sending.

Voice command overlay

While recording, a full-screen overlay shows:

• Live waveform — real-time audio visualization
• Recording duration — seconds elapsed
• Cancel hint — swipe away to cancel

Speech-to-text

The app uses the device’s built-in speech recognition:

Platform	Engine
iOS	Apple Speech Recognition
Android	Google Speech Recognition

No API keys are required — transcription happens on-device or through the platform’s cloud service.

Tips for better recognition

1. Speak clearly and at a natural pace
2. Minimize background noise — find a quiet environment
3. Hold the phone 6–12 inches from your mouth
4. Use short phrases for best accuracy
5. Set your device language to match the language you’re speaking

Limitations

• Voice input requires microphone permission (granted during setup)
• Recording requires a physical device — not available in simulators
• Maximum recording duration: 60 seconds per message
• Transcription accuracy depends on your device’s speech engine and language support
• Text-to-speech (TTS) for agent responses is currently a desktop-only feature

Troubleshooting

Microphone icon not responding

1. Check microphone permission in device Settings > Snippbot
2. Ensure no other app is using the microphone
3. Try restarting the app

Transcription is inaccurate

1. Speak more slowly and clearly
2. Check your device language settings match your spoken language
3. Reduce background noise
4. Try shorter phrases