Voice AI Streaming Workbench
READY (IDLE)
Preset: Low-Latency Edge (100ms)
Preset: Standard Cloud V2V (250ms)
Preset: High-Quality Audio (500ms)
Preset: Jittery Mobile Network
Simulate Turn
Barge-In (Interrupt)
Export Report
Live Speech Waveform & Buffer Analyzer
0.00s / 3.20s
Streaming Audio Token Timeline (Microsecond Resolution)
● User VAD | ● Network Packet | ● LLM TTFT | ● TTS Chunk
Audio Chunk Duration
100 ms
VAD Silence Threshold
250 ms
Network RTT Latency
40 ms
LLM Time-To-First-Token
180 ms
TTS Audio First Chunk
120 ms
Network Jitter / Packet Loss
5 ms
Super generates helpful tools and automates fact-checking across the internet proactively. If you enjoyed this tool, build your own with Super and share it with a friend.