Simuliert eine Konversation

Deprecated

POST

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

POST

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

curl -X POST https://api.elevenlabs.io/v1/convai/agents/agent_3701k3ttaq12ewp8b7qv5rfyszkz/simulate-conversation \
     -H "Content-Type: application/json" \
     -d '{
  "simulation_specification": {
    "simulated_user_config": {
      "first_message": "Hello, how can I help you today?",
      "language": "en",
      "disable_first_message_interruptions": false
    }
  }
}'

{
  "simulated_conversation": [
    {
      "role": "user",
      "time_in_call_secs": 1,
      "agent_metadata": {
        "agent_id": "string",
        "branch_id": "string",
        "workflow_node_id": "string",
        "version_id": "string"
      },
      "message": "string",
      "multivoice_message": {
        "parts": [
          {
            "text": "string",
            "voice_label": "string",
            "time_in_call_secs": 1
          }
        ]
      },
      "tool_calls": [
        {
          "request_id": "string",
          "tool_name": "string",
          "params_as_json": "string",
          "tool_has_been_called": true,
          "type": "system",
          "tool_details": {
            "type": "webhook",
            "method": "string",
            "url": "string",
            "body": "string",
            "headers": {},
            "path_params": {},
            "query_params": {}
          }
        }
      ],
      "tool_results": [
        {
          "dynamic_variable_updates": [
            {
              "new_value": "string",
              "old_value": "string",
              "tool_name": "string",
              "tool_request_id": "string",
              "updated_at": 1.1,
              "variable_name": "string"
            }
          ],
          "error_type": "",
          "is_blocked": false,
          "is_error": true,
          "raw_error_message": "",
          "request_id": "string",
          "result_value": "string",
          "tool_has_been_called": true,
          "tool_latency_secs": 0,
          "tool_name": "string",
          "type": "client"
        }
      ],
      "feedback": {
        "score": "like",
        "time_in_call_secs": 1
      },
      "llm_override": "string",
      "producing_llm": "string",
      "conversation_turn_metrics": {
        "metrics": {},
        "convai_asr_provider": "string",
        "convai_tts_model": "string",
        "convai_tts_cascade": "string"
      },
      "rag_retrieval_info": {
        "chunks": [
          {
            "document_id": "string",
            "chunk_id": "string",
            "vector_distance": 1.1
          }
        ],
        "embedding_model": "e5_mistral_7b_instruct",
        "retrieval_query": "string",
        "rag_latency_secs": 1.1,
        "used_chunk_ids": [
          "string"
        ]
      },
      "llm_usage": {
        "model_usage": {}
      },
      "interrupted": false,
      "ignored_as_backchannel": false,
      "original_message": "string",
      "reasoning": [
        {
          "summary": "string",
          "provider_redact": false
        }
      ],
      "source_medium": "audio",
      "source_event_id": 1,
      "used_static_kb_document_ids": [
        "string"
      ],
      "user_identifier": "string",
      "id": "string",
      "triggered_guardrails": [
        {
          "guardrail_type": "custom",
          "guardrail_name": "string"
        }
      ],
      "file_input": {
        "file_id": "string",
        "original_filename": "string",
        "mime_type": "string",
        "file_url": "string"
      },
      "file_inputs": [
        {
          "file_id": "string",
          "original_filename": "string",
          "mime_type": "string",
          "file_url": "string"
        }
      ],
      "contextual_update_info": {
        "context_id": "string",
        "is_superseded": false
      },
      "reasoned": false
    }
  ],
  "analysis": {
    "call_successful": "success",
    "transcript_summary": "string",
    "evaluation_criteria_results": {},
    "data_collection_results": {},
    "evaluation_criteria_results_list": [
      {
        "criteria_id": "string",
        "result": "success",
        "rationale": "string",
        "scoring_mode": "binary",
        "score": 1,
        "max_score": 1
      }
    ],
    "data_collection_results_list": [
      {
        "data_collection_id": "string",
        "rationale": "string",
        "name": "string",
        "value": null,
        "json_schema": {
          "type": "string",
          "description": "A user-provided message"
        }
      }
    ],
    "call_success_score": 1.1,
    "call_summary_title": "string",
    "scoped": [
      {
        "scope": "conversation",
        "source_agent_id": "string",
        "successful": "success",
        "source_branch_id": "string",
        "evaluation_criteria_results": {},
        "data_collection_results": {},
        "success_score": 1.1
      }
    ]
  }
}

Veraltet. Verwenden Sie die Endpunkte /v1/convai/agent-testing/create und /v1/convai/agents/:agent_id/run-tests, um Simulationen zu erstellen und auszuführen. Führen Sie ein Gespräch zwischen dem Agenten und einem simulierten Benutzer durch.

Pfadparameter

agent_idstringErforderlich

Die ID eines Agenten. Sie wird beim Erstellen des Agenten zurückgegeben.

Anfrage

This endpoint expects an object.

simulation_specificationobjectErforderlich

Eine Spezifikation, die beschreibt, wie das Gespräch simuliert werden soll

extra_evaluation_criterialist of objects or nullOptional

Eine Liste der zu testenden Bewertungskriterien

new_turns_limitintegerOptionalStandardwert ist 10000

Maximale Anzahl neuer Gesprächswechsel, die in der Gesprächssimulation generiert werden sollen

Antwort

Erfolgreiche Antwort

simulated_conversationlist of objects

analysisobject

Fehler

422

Unprocessable Entity Error

Pfadparameter

Header

Anfrage

Antwort

Fehler