Latency of a streaming voice pipeline
Voice agent
- The serial turn
- Stream every stage
- Stop when the caller cuts in
- A pipeline in each region
- Final test
A phone agent listens, thinks and speaks. Build the chain of speech to text, model and text to speech, stream every stage, stop talking when the caller cuts in, and place a pipeline in each region.
Final testThe model slows 1.4 times for 20 s at 0:40. Your final board runs once more with a new traffic seed.
Stage 4 of 4 · 8 parts given · 16 requests/s
Progress on this page stays on this device.