Voxtral Pipeline Tester
Configuration
TTS Endpoint (HTTP)
STT WebSocket
Audio Capture + Realtime STT
OFF
WS OFF
Source
Microphone
Browser Tab (Screen Share)
Start Capture + STT
Stop
Force Commit
Audio Level
Live Transcript
Copy
(waiting for speech...)
Text-to-Speech
Text
Hello, this is a test of the Voxtral text to speech pipeline.
Voice
Synthesize
Stop Playback
Full Pipeline (Realtime STT -> LLM -> TTS)
LLM System Prompt
You are a helpful voice assistant. Keep responses brief and conversational.
Worker URL (for LLM)
Start Pipeline
Stop
Event Log