Simulate conversation

POST

/v1/convai/agents/:agent_id/simulate-conversation

POST

/v1/convai/agents/:agent_id/simulate-conversation

1 from elevenlabs import (
2     AgentConfig,
3     ConversationSimulationSpecification,
4     ElevenLabs,
5 )
6 
7 client = ElevenLabs(
8     api_key="YOUR_API_KEY",
9 )
10 client.conversational_ai.agents.simulate_conversation(
11     agent_id="21m00Tcm4TlvDq8ikWAM",
12     simulation_specification=ConversationSimulationSpecification(
13         simulated_user_config=AgentConfig(
14             first_message="Hello, how can I help you today?",
15             language="en",
16         ),
17     ),
18 )

Try it

200Successful

1 {
2   "simulated_conversation": [
3     {
4       "role": "user",
5       "time_in_call_secs": 42,
6       "message": "foo",
7       "tool_calls": [
8         {
9           "request_id": "foo",
10           "tool_name": "foo",
11           "params_as_json": "foo",
12           "tool_has_been_called": true,
13           "type": "foo",
14           "tool_details": {
15             "type": "webhook",
16             "method": "foo",
17             "url": "foo",
18             "headers": {},
19             "path_params": {},
20             "query_params": {},
21             "body": "foo"
22           }
23         }
24       ],
25       "tool_results": [
26         {
27           "request_id": "foo",
28           "tool_name": "foo",
29           "result_value": "foo",
30           "is_error": true,
31           "tool_has_been_called": true,
32           "type": "foo",
33           "tool_latency_secs": 0
34         }
35       ],
36       "feedback": {
37         "score": "like",
38         "time_in_call_secs": 42
39       },
40       "llm_override": "foo",
41       "conversation_turn_metrics": {
42         "metrics": {}
43       },
44       "rag_retrieval_info": {
45         "chunks": [
46           {
47             "document_id": "foo",
48             "chunk_id": "foo",
49             "vector_distance": 42
50           }
51         ],
52         "embedding_model": "e5_mistral_7b_instruct",
53         "retrieval_query": "foo",
54         "rag_latency_secs": 42
55       },
56       "llm_usage": {
57         "model_usage": {}
58       },
59       "interrupted": false,
60       "original_message": "foo",
61       "source_medium": "audio"
62     }
63   ],
64   "analysis": {
65     "call_successful": "success",
66     "transcript_summary": "foo",
67     "evaluation_criteria_results": {},
68     "data_collection_results": {}
69   }
70 }

Run a conversation between the agent and a simulated user.

Path parameters

agent_idstringRequired

The id of an agent. This is returned on agent creation.

Request

This endpoint expects an object.

simulation_specificationobjectRequired

A specification detailing how the conversation should be simulated

extra_evaluation_criterialist of objects or nullOptional

A list of evaluation criteria to test

Response

Successful Response

simulated_conversationlist of objects

analysisobject

Simulate conversation

Path parameters

Headers

Request

Response

Errors