[Maxim] Voice simulation runs docs

Author: impoiler

docs.json

@@ -190,7 +190,8 @@
                             {
                                 "group": "Voice Simulation",
                                 "pages": [
-                                    "simulations/voice-simulation/voice-simulation"
+                                    "simulations/voice-simulation/voice-simulation",
+                                    "simulations/voice-simulation/simulation-runs"
                                 ]
                             }
                         ]

images/docs/evaluate/how-to/evaluate-workflows-via-api-endpoint/evaluate-simulated-voice-agent/single-entry-result.png

@@ -7,7 +7,7 @@
 		},
 		"voice-simulation": {
 			"title": "Voice Simulation",
-			"pages": ["voice-simulation"]
+			"pages": ["voice-simulation", "simulation-runs"]
 		}
 	}
 }

images/docs/evaluate/how-to/evaluate-workflows-via-api-endpoint/evaluate-simulated-voice-agent/trigger-voice-simulation-testrun.png

@@ -0,0 +1,58 @@
+---
+title: Voice Simulation Runs
+description: Test your Voice Agent's interaction capabilities with realistic voice simulations on thousands of scenarios and evaluate your agent.
+---
+
+## Evaluate voice agent performance with simulated sessions at scale with multiple schenarios with test runs.
+
+You can run test runs with dataset containing multiple scenarios for your voice agent,
+
+<Steps>
+
+<Step title="Create a Dataset for testing">
+- Configure the agent dataset template with:
+- **Agent scenarios**: Define specific situations for testing (e.g., "Update address", "Order an iPhone")
+- **Expected steps**: List expected actions and responses
+
+![Voice Agent Dataset](/images/docs/evaluate/how-to/evaluate-workflows-via-api-endpoint/evaluate-simulated-voice-agent/voice-agent-dataset.png)
+</Step>
+
+<Step title="Set up the Test Run">
+- Navigate to your voice agent, click "Test", and "Simulated session" mode will be pre-selected since voice agent can not be tested single turn.
+- Pick your agent dataset from the dropdown
+- Select relevant evaluators
+
+<Note>
+  At the moment, only built-in evaluators are supported for voice simulation runs; custom evaluators will be supported in the future.
+</Note>
+
+![Configure simulation Test Run](/images/docs/evaluate/how-to/evaluate-workflows-via-api-endpoint/evaluate-simulated-voice-agent/trigger-voice-simulation-testrun.png)
+
+</Step>
+
+<Step title="Execute Test Run">
+- Click "Trigger test run" to begin
+- The system will call to your voice agent and simulate conversations for each scenario
+</Step>
+
+<Step title="Review results">
+- Each session runs end-to-end for thorough evaluation
+- You'll see detailed results for every scenario
+- Text based evaluators will be evaluated via turns of the call transcription rest will be evaluated by call recording audio file.
+
+![Simulation Test Run result](/images/docs/evaluate/how-to/evaluate-workflows-via-api-endpoint/evaluate-simulated-voice-agent/voice-simulation-testrun-report.png)
+</Step>
+
+<Step title="Inspect the single entry">
+- You can click on any entry to inspect the detailed results for that specific scenario.
+- By default, test run will evalouate few metrics for each entry based on the recording audio file which are:
+  - **Avg Latency**: How long it took the agent to respond to the prompt
+  - **Talk ratio**: How much of the time the agent was talking compared to the time the simulation agent was talking
+  - **Avg pitch**: The average pitch of the agent's response
+  - **Words per minute**: How many words the agent was able to speak in a minute
+
+![Simulation Test Run result](/images/docs/evaluate/how-to/evaluate-workflows-via-api-endpoint/evaluate-simulated-voice-agent/single-entry-result.png)
+</Step>
+
+</Steps>
+