Description
SubVox is a voice-activated content extraction tool with optional conversational AI features, designed for creators, performers, productivity users, and anyone who needs fast, hands-free action triggers. Speak your start and stop trigger phrases, and SubVox interprets the intent — then sends it to the third-party app or integration you choose.
All audio and speech recognition is performed locally on your device by default. You may optionally choose to turn on Cloud Speech AI, which additionally can use AI provider cloud models for speech to text. SubVox does not store, track, or collect audio, transcripts, or extracted text.
Features
• Voice-triggered context extraction
• Custom start and stop phrases
• On-device speech processing
• Optional conversational AI using user-provided AI provider credentials
• Optional background mode when the screen is off
• Third-party integrations including MysterSmith, Inject, and more
• watchOS companion app
• Secure storage of integration and AI credentials in Keychain
Privacy First
SubVox does not store audio, transcripts, or extracted text.
If you choose to enable conversational AI, text is sent directly from your device to the AI provider you select using your own credentials. SubVox does not receive, store, or process this data on its own servers.
Your data only leaves your device when you explicitly choose to send it and only to services you authorize.
For Entertainment & Personal Utility
SubVox is intended for entertainment and personal utility. It is not designed for professional, legal, medical, or safety-critical use.
What's new (v1.0.6)
- bug fixes and usabilty improvements