Skip to main content

Documentation Index

Fetch the complete documentation index at: https://docs.usetuner.ai/llms.txt

Use this file to discover all available pages before exploring further.

​
What is Call Simulation?

Testing voice AI manually is slow, incomplete, and nobody’s favourite task. Call Simulation makes actual voice calls to your agent so you know it’s ready, before launch, after every update, and every time something changes. You define what your agent does. Tuner builds the scenarios that stress test it, dials in over SIP, and scores every call against the same Evals you’ve configured for production. The simulation landing page before your first run

​
How It Works

1

Connect your agent

Configure the SIP endpoint Tuner uses to reach your voice agent. This is a one-time setup in Agent Settings → SIP Settings. The connection method depends on your voice AI provider.SIP configuration in Agent Settings
2

Configure your Simulation Mix

Choose the number of calls (5–20 per batch) and set the ratio of Routine Callers to Pressure Tests. You can also narrow which behaviour checks the Pressure Tests target.Run Simulation dialog, set count, mix, and target checks
3

Select Pressure Test targets

By default, Pressure Tests target every behaviour check. Narrow them to specific failure modes, hallucination, scope boundary, escalation handling, offensive language, personal data, and more. Unselected checks won’t generate scenarios.Choose which behaviour checks Pressure Tests should target
4

Review results

Every completed call appears in the Simulations table with type, intent, outcome, tags, sentiment, ended reason, duration, and cost. Click any row for the full transcript, latency breakdown, tool traces, and evaluation results.Simulation results, each call shows type, intent, outcome, and sentiment

​
The Simulation Mix

You control what gets tested. Run a full health check or narrow down to a single Intent, Call Outcome, or behaviour check, every run is yours to configure.
Routine CallersPressure Tests
What they doAI Callers that simulate your everyday customers, cooperative, on-topic, working through your agent’s normal workflowsAI Callers that simulate the difficult ones, unhappy customers, edge cases, people trying to bypass your process or ask for something outside scope
What you learnHow your agent performs on the happy path when everything goes the way it’s supposed toWhere your agent has a breaking point, hallucination, scope boundary, escalation handling, and other failure modes
How they’re generatedFrom your agent’s configured Intents and Call OutcomesAuto-generated from your configured behaviour checks

​
Why volume matters

In voice, the same intent said ten different ways produces ten different conversations. Rushed, hesitant, overly detailed, unclear, small differences in how a caller speaks can completely change how your agent responds. The more simulations you run, the more of these variations you cover.

​
Your Production Evals Run Every Simulation

No separate scoring. No extra configuration. Every simulation call is evaluated against the exact same Evals, Call Outcomes, and Intents you’ve already set up for your live agent.
  • Same Evals, every run. Whatever you’ve configured to catch failures in production, hallucinations, missed Intents, wrong Call Outcomes, runs automatically on every simulation call. What fails in production will fail here first.
  • See failures before your customers do. Every run surfaces exactly what passed and what failed. Fix the failures, adjust your agent, and run again.
  • Regression built-in. Changed a prompt? Updated a workflow? Every time you run a simulation, you’re also checking that nothing you already fixed has broken again.

​
Getting Started

Connect your voice agent to Tuner via SIP. The setup depends on your provider:

Vapi Setup Guide

Create a SIP trunk in Vapi, configure credentials in Tuner, and link your assistant.

Retell Setup Guide

No SIP trunk needed, enter the Retell SIP URI in Tuner and run your first batch.

LiveKit Setup Guide

Create a LiveKit inbound trunk and dispatch rule, then wire SIP correlation into your agent.

​
FAQ

Each simulated call takes 1–4 minutes depending on scenario complexity. A batch of 10 calls typically completes in 5–10 minutes.
Yes. Each simulated call consumes credits from your balance. The rate is shown in the Run Simulation dialog before you start.
Yes. Simulation calls are separate from production traffic. They’re tagged as SIMULATED and don’t affect your production metrics.
Vapi, Retell, and LiveKit are currently supported. More providers will be added. If your provider supports SIP, you can likely connect it, contact support for guidance.