Emergency dispatch triage
A caller in distress will talk over a slow bot. The agent has to yield instantly, keep the address and the symptom in memory, and never replay a half-sentence the caller already corrected.
Case study ยท NerdWare Tech
A streaming voice agent built to answer inside 500 ms and go silent the instant you cut in.
State: LISTENING
VAD armed. Waiting for an agent turn.
Each stage starts streaming before the last one finishes. The legacy cascade waits for whole files at every hop.
Target budget / illustrative. Cascade figures are a representative range for a record, upload, batch-transcribe, complete, synthesize bot (about 1.5 to 2.5 s), not a benchmark of any product.
Audio frames move through memory from the caller's mic to the caller's speaker. Nothing is written to disk and nothing waits on a poll.
When the VAD hears the caller during agent speech, the worker does four things in one tick, in this order.
Stateful memory sits beside the LLM: the last turns plus retrieved facts for this caller, fetched while STT is still finalizing.
The agent emits four timings per turn, so a slow stage is visible the same hour it regresses. The rows below are simulated by the scope above.
| Turn | Transport | STT | LLM TTFT | TTS TTFB | Round trip | Barge-in |
|---|
A caller in distress will talk over a slow bot. The agent has to yield instantly, keep the address and the symptom in memory, and never replay a half-sentence the caller already corrected.
Long, interrupted narratives. Barge-in lets the caller fix a date mid-sentence without restarting the form.
Natural turn-taking matters more than polish. A reply that starts under half a second feels like a person, not a queue.
NerdWare Tech
Tell us the calls you want handled. The waveform above is a simulation and the latency figures are target budgets.