Voice performance
PlannedLatency & Barge-in Lab
A proposed measurement rig for where the seconds go in a voice turn, and how cleanly the agent handles being interrupted.
01 / THE PROBLEM
Why this matters.
Voice agents feel wrong long before they answer wrong. Silence after the caller stops speaking, or an agent that talks over an interruption, breaks the conversation regardless of how good the answer was.
02 / THE DIRECTION
The proposed solution.
A rig that instruments each stage of a turn — endpoint detection, model first token, speech first byte, network — so the slow stage is identifiable rather than inferred. Alongside it, a barge-in set that interrupts the agent at defined points and checks that it stops, keeps its turn state, and resumes coherently.
03 / ARCHITECTURE
How the pieces connect.
Conceptual flow — proposed architecture
- 01Instrument the turn
- 02Endpointing, first token, first byte
- 03Budget per stage
- 04Barge-in test set
- 05Interruption behaviour check
- 06Tuning record
04 / ENGINEERING DECISIONS
Designed with intent.
- Measure per stage, because one end-to-end number hides which component to fix.
- Set a budget per stage and treat exceeding it as a defect rather than a preference.
- Treat barge-in as correctness, not polish — the agent must stop and keep its state.
- Record what each tuning change cost elsewhere, since latency and answer quality trade against each other.
05 / EVALUATION APPROACH
What needs to hold true.
- Test interruption at the start, middle, and end of an agent turn.
- Test a caller who pauses mid-sentence and expects to keep the turn.
- Compare measured stages against the budget on every configuration change.
- Check that latency tuning has not regressed the behaviour checks in the regression harness.