Gespreksevaluatie uitvoeren

POST

/v1/convai/conversations/:conversation_id/analysis/evaluations/run

POST

/v1/convai/conversations/:conversation_id/analysis/evaluations/run

curl -X POST https://api.elevenlabs.io/v1/convai/conversations/conversation_id/analysis/evaluations/run \
     -H "Content-Type: application/json" \
     -d '{
  "evaluation_id": "evaluation_id"
}'

{
  "agent_id": "agent_3701k3ttaq12ewp8b7qv5rfyszkz",
  "status": "processing",
  "metadata": {
    "start_time_unix_secs": 1714423232,
    "call_duration_secs": 10,
    "accepted_time_unix_secs": 1,
    "queue_wait_secs": 1.1,
    "cost": 1,
    "deletion_settings": {
      "deletion_time_unix_secs": 1,
      "deleted_logs_at_time_unix_secs": 1,
      "deleted_audio_at_time_unix_secs": 1,
      "deleted_transcript_at_time_unix_secs": 1,
      "delete_transcript_and_pii": true,
      "delete_audio": true
    },
    "feedback": {
      "type": "thumbs",
      "overall_score": "like",
      "likes": 1,
      "dislikes": 1,
      "rating": 1,
      "comment": "comment"
    },
    "authorization_method": "invalid",
    "charging": {
      "dev_discount": true,
      "is_burst": true,
      "tier": "tier",
      "llm_price": 1.1,
      "llm_charge": 1,
      "call_charge": 1,
      "platform_charge": 1,
      "platform_price": 1.1,
      "free_minutes_consumed": 1.1,
      "free_llm_dollars_consumed": 1.1,
      "analysis": {
        "total": {},
        "last_run": {}
      },
      "voice_surcharge_charge": 1
    },
    "phone_call": {
      "type": "exotel",
      "agent_number": "agent_number",
      "call_sid": "call_sid",
      "direction": "inbound",
      "external_number": "external_number",
      "phone_number_id": "phone_number_id",
      "stream_sid": "stream_sid",
      "call_result": "answered"
    },
    "batch_call": {
      "batch_call_id": "batch_call_id",
      "batch_call_recipient_id": "batch_call_recipient_id",
      "campaign": {
        "campaign_id": "campaign_id",
        "campaign_lead_id": "campaign_lead_id"
      }
    },
    "termination_reason": "termination_reason",
    "error": {
      "code": 1,
      "reason": "reason"
    },
    "warnings": [
      "warnings"
    ],
    "main_language": "main_language",
    "rag_usage": {
      "usage_count": 1,
      "embedding_model": "embedding_model"
    },
    "text_only": true,
    "features_usage": {
      "pii_zrm_workspace": true,
      "pii_zrm_agent": true,
      "is_livekit": true
    },
    "eleven_assistant": {
      "is_eleven_assistant": true
    },
    "initiator_id": "initiator_id",
    "conversation_initiation_source": "unknown",
    "conversation_initiation_source_version": "conversation_initiation_source_version",
    "timezone": "timezone",
    "async_metadata": {
      "delivery_status": "pending",
      "delivery_timestamp": 1,
      "external_system": "external_system",
      "external_id": "external_id",
      "delivery_error": "delivery_error",
      "external_link": "external_link",
      "retry_count": 1,
      "last_retry_timestamp": 1,
      "last_processed_external_message_id": "last_processed_external_message_id"
    },
    "whatsapp": {
      "whatsapp_user_id": "whatsapp_user_id",
      "direction": "inbound",
      "whatsapp_phone_number_id": "whatsapp_phone_number_id",
      "awaiting_first_user_message": true,
      "call_permission_request": {
        "state": "blocked"
      },
      "call_result": "answered",
      "opening_message": {
        "template_name": "template_name",
        "sent_at_unix": 1
      }
    },
    "sms": {
      "direction": "inbound",
      "sms_user_phone_number": "sms_user_phone_number",
      "phone_number_id": "phone_number_id",
      "agent_phone_number": "agent_phone_number"
    },
    "agent_created_from": "cli",
    "agent_last_updated_from": "cli",
    "voice_rewards": [
      {
        "voice_id": "voice_id",
        "reward_usd_cents": 1.1
      }
    ],
    "cost_fiat": 1.1
  },
  "conversation_id": "conv_7401k5m9x2p8ec3rqv6dtnhb0fzw",
  "has_audio": true,
  "has_user_audio": true,
  "has_response_audio": true,
  "has_auxiliary_audio": true,
  "transcript": [
    {
      "role": "user",
      "time_in_call_secs": 10,
      "agent_metadata": {
        "agent_id": "agent_id"
      },
      "message": "Hello, how are you?",
      "multivoice_message": {
        "parts": [
          {
            "text": "text",
            "voice_label": null,
            "time_in_call_secs": null
          }
        ]
      },
      "tool_calls": [
        {
          "request_id": "request_id",
          "tool_name": "tool_name",
          "params_as_json": "params_as_json",
          "tool_has_been_called": true
        }
      ],
      "tool_results": [
        {
          "is_error": true,
          "request_id": "request_id",
          "result_value": "result_value",
          "tool_has_been_called": true,
          "tool_name": "tool_name"
        }
      ],
      "feedback": {
        "score": "like",
        "time_in_call_secs": 1
      },
      "llm_override": "llm_override",
      "producing_llm": "producing_llm",
      "rag_retrieval_info": {
        "chunks": [
          {
            "document_id": "document_id",
            "chunk_id": "chunk_id",
            "vector_distance": 1.1
          }
        ],
        "embedding_model": "e5_mistral_7b_instruct",
        "retrieval_query": "retrieval_query",
        "rag_latency_secs": 1.1
      },
      "interrupted": true,
      "ignored_as_backchannel": true,
      "platform_event": "user_turn_timeout",
      "original_message": "original_message",
      "reasoning": [
        {}
      ],
      "source_medium": "audio",
      "source_event_id": 1,
      "used_static_kb_document_ids": [
        "used_static_kb_document_ids"
      ],
      "user_identifier": "user_identifier",
      "id": "id",
      "triggered_guardrails": [
        {
          "guardrail_type": "custom"
        }
      ],
      "file_input": {
        "file_id": "file_id",
        "original_filename": "original_filename",
        "mime_type": "mime_type",
        "file_url": "file_url"
      },
      "file_inputs": [
        {
          "file_id": "file_id",
          "original_filename": "original_filename",
          "mime_type": "mime_type",
          "file_url": "file_url"
        }
      ],
      "contextual_update_info": {
        "context_id": "context_id"
      },
      "reasoned": true
    }
  ],
  "agent_name": "My agent",
  "conversation_product": "conversation_product",
  "user_id": "user_id",
  "branch_id": "branch_id",
  "version_id": "agtvrsn_5xM3yVvZQKV0EfqQpLr2",
  "analysis": {
    "call_successful": "success",
    "transcript_summary": "transcript_summary",
    "evaluation_criteria_results": {
      "key": {
        "criteria_id": "criteria_id",
        "result": "success",
        "rationale": "rationale"
      }
    },
    "data_collection_results": {
      "key": {
        "data_collection_id": "data_collection_id",
        "rationale": "rationale",
        "json_schema": {
          "type": "string",
          "description": "A user-provided message"
        }
      }
    },
    "evaluation_criteria_results_list": [
      {
        "criteria_id": "criteria_id",
        "result": "success",
        "rationale": "rationale"
      }
    ],
    "data_collection_results_list": [
      {
        "data_collection_id": "data_collection_id",
        "rationale": "rationale",
        "json_schema": {
          "type": "string",
          "description": "A user-provided message"
        }
      }
    ],
    "call_success_score": 1.1,
    "call_summary_title": "call_summary_title",
    "scoped": [
      {
        "scope": "conversation",
        "source_agent_id": "source_agent_id",
        "successful": "success"
      }
    ]
  },
  "user_message_count": 1,
  "agent_message_count": 1,
  "visited_agents": [
    {
      "agent_id": "agent_id",
      "branch_id": "branch_id"
    }
  ],
  "conversation_initiation_client_data": {
    "conversation_config_override": {
      "asr": {
        "keywords": [
          "hello",
          "world"
        ]
      },
      "turn": {
        "soft_timeout_config": {
          "message": "Hhmmmm...yeah."
        }
      },
      "tts": {
        "model_id": "eleven_turbo_v2",
        "voice_id": "cjVigY5qzO86Huf0OWal",
        "stability": 0.5,
        "speed": 1,
        "similarity_boost": 0.8,
        "pronunciation_dictionary_locators": [
          {
            "pronunciation_dictionary_id": "pronunciation_dictionary_id",
            "version_id": null
          }
        ]
      },
      "conversation": {
        "max_duration_seconds": 600
      },
      "agent": {
        "first_message": "Hello, how can I help you today?",
        "language": "en",
        "prompt": {
          "prompt": "You are a helpful assistant that can answer questions about the topic of the conversation.",
          "llm": "gemini-2.0-flash-001",
          "tool_ids": [
            "tool_ids"
          ],
          "knowledge_base": [
            {
              "type": "file",
              "name": "My Knowledge Base",
              "id": "123",
              "usage_mode": "auto"
            }
          ]
        }
      }
    },
    "custom_llm_extra_body": {
      "key": "value"
    },
    "user_id": "user_id",
    "source_info": {
      "source": "unknown",
      "version": "version"
    },
    "branch_id": "branch_id",
    "environment": "environment",
    "starting_workflow_node_id": "starting_workflow_node_id",
    "procedure_ids": [
      "procedure_ids"
    ],
    "start_procedure_ids": [
      "start_procedure_ids"
    ],
    "dynamic_variables": {
      "key": "value"
    }
  },
  "environment": "production",
  "tag_ids": [
    "tag_ids"
  ],
  "otlp_traces": {
    "key": "value"
  }
}

Voer een specifieke evaluatie voor een gesprek opnieuw uit.

Padparameters

conversation_idstringVerplicht

ID van het gesprek

Verzoek

This endpoint expects an object.

evaluation_idstringVerplicht

ID van het afzonderlijke evaluatiecriterium dat opnieuw moet worden uitgevoerd.

scopeenumOptioneelStandaard conversation

Toegestane waarden:

Response

Geslaagde response

agent_idstring

statusenum

Toegestane waarden:

metadataobject

conversation_idstring

has_audioboolean

has_user_audioboolean

has_response_audioboolean

has_auxiliary_audioboolean

transcriptlist of objects

agent_namestringOptioneel

conversation_productstringOptioneelStandaard agent

user_idstringOptioneel

branch_idstringOptioneel

version_idstringOptioneel

De ID van de agentversie die voor dit gesprek is gebruikt

analysisobjectOptioneel

user_message_countintegerOptioneel

Aantal beurten in het gebruikerstranscript met tekst; beurten met alleen tools worden niet meegeteld. null voor oudere gesprekken van vóór dit veld waarvan het transcript is verwijderd.

agent_message_countintegerOptioneel

Aantal beurten in het agenttranscript met tekst; beurten met alleen tools worden niet meegeteld. null voor oudere gesprekken van vóór dit veld waarvan het transcript is verwijderd.

visited_agentslist of objectsOptioneel

conversation_initiation_client_dataobjectOptioneel

environmentstringOptioneelStandaard production

tag_idslist of stringsOptioneel

ID's van gesprekstags die aan dit gesprek zijn toegewezen.

otlp_tracesmap from strings to anyOptioneel

OpenTelemetry-tracepayload wanneer de request format=opentelemetry gebruikt; anders weggelaten.

Fouten

422

Analysis Run Evaluation Request Unprocessable Entity Error

Padparameters

Headers

Verzoek

Response

Fouten