Voice Integration
Voice Integration
V1 Documentation - Under Heavy Rework. Some sections may be incomplete. For the latest info, check the FAB listing or Discord.
Talk to the plugin with your voice and hear AI responses spoken back.
Speech-to-Text (STT)
Record your voice and have it transcribed to text using OpenAI Whisper.
Setup
- Go to Settings > Voice & Sound
- Enter your OpenAI API key (used for Whisper transcription)
- Enable “Voice Enabled”
How to Use
- Click the Voice button in the Architect chat toolbar
- Speak your prompt
- The transcription appears in the text input
- Review and send, or enable auto-send
Auto-Send Transcription
When enabled, the transcribed text is sent automatically without manual confirmation. Toggle in Settings > Voice & Sound.
Text-to-Speech (TTS)
Hear AI responses read aloud using OpenAI TTS.
Available Voices
| Voice | Description |
|---|---|
| Alloy | Neutral, balanced |
| Echo | Male |
| Fable | British male |
| Onyx | Deep male |
| Nova | Female |
| Shimmer | Soft female |
Configuration
- Go to Settings > Voice & Sound
- Select your preferred voice from the “TTS Voice (Auto-Play)” dropdown
- The voice setting persists across sessions
Auto-Play Responses
When enabled, every AI response is automatically spoken aloud after it finishes streaming. Toggle in Settings > Voice & Sound > “Auto-Play Responses”.
Requirements
- An OpenAI API key is required for both STT (Whisper) and TTS
- A working microphone for voice recording
- The Voice API key can be set separately from your main AI provider key