---
title: "Import Conversation"
method: POST
path: "/conversations/import"
tags: ["Conversation"]
---

# Import Conversation

`POST /conversations/import`

Import a conversation from a media file. This endpoint creates a new conversation from an external source (e.g., Gong, Salesforce) and processes it through the Attention platform's transcription and analysis pipeline.

## Request body

- ImportConversationRequest
  - `mediaURL` string, required — URL of the media file to import (audio or video)
  - `userID` string, required — UUID of the user who owns this conversation
  - `applicationName` string — Name of the external application (e.g., 'zap:gong', 'salesforce')
  - `applicationExternalID` string — External ID from the source application
  - `conversationTitle` string — Title or name of the conversation
  - `conversationStartedAt` string, date-time — When the conversation started (ISO 8601 format)
  - `headers` Header[] — HTTP headers to use when fetching the media file
    - `key` string, required — Header name
    - `value` string, required — Header value
  - `opportunityID` string — ID of the associated opportunity or deal
  - `linkedCrmRecords` LinkedCrmRecord[] — CRM records linked to this conversation (accounts, contacts, etc.)
    - `id` string, required
    - `code` string, required
  - `skipScorecardCalculation` boolean — Skip automatic scorecard calculation for this conversation
  - `skipCRMFieldsCalculation` boolean — Skip CRM fields calculation for this conversation
  - `skipOpportunitiesExport` boolean — Skip exporting opportunities for this conversation
  - `transcriptionSettings` TranscriptionSettings
    - `gladia` GladiaTranscriptionSettings
      - `enableDiarization` boolean — Enable speaker diarization (identifying different speakers)
  - `transcript` TranscriptInput
    - `v1` TranscriptV1Input
      - `pieceInFlux` string, required — Current incomplete piece being transcribed
      - `final` TranscriptV1Item[], required — Finalized transcript items
        - `speaker` string — Speaker identifier
        - `sentence` string — Transcribed text
        - `startTimestamp` number, float, required — Start time in seconds
        - `endTimestamp` number, float — End time in seconds
    - `v2` TranscriptV2Input
      - `pieceInFlux` string, required — Current incomplete piece being transcribed
      - `final` TranscriptV2Item[], required — Finalized transcript items with word-level timing
        - `speaker` string, required — Speaker identifier
        - `words` Word[], required — Individual words with timestamps
          - `text` string, required — Word text
          - `startTimestamp` number, float, required — Start time in seconds
          - `endTimestamp` number, float, required — End time in seconds
  - `externalMetadata` ExternalMetadata
    - `participants` ExternalMetadataParticipant[] — List of conversation participants
      - `email` string, required — Participant email address
      - `name` string, required — Participant name
      - `accessType` 'Internal' | 'External', required — Whether the participant is internal or external to the organization
      - `organizer` boolean, required — Whether this participant is the organizer of the conversation
    - `extra` object — Additional metadata as key-value pairs

## Response `200`

Conversation successfully imported

- ImportConversationResponse
  - `uuid` string, required — UUID of the newly imported conversation

## Other responses

- `400` — Invalid request parameters or user does not belong to organization
- `401` — Authentication failed - valid API key required
- `403` — User does not belong to the organization
- `default` — Unexpected error occurred

---

[API](https://skmtc.net/attention/apis/attention-service-v2.md) · [All operations](https://skmtc.net/attention/apis/attention-service-v2/llms.txt) · [OpenAPI document](https://skmtc-service-staging.skmtc.workers.dev/v1/apis/attention/attention-service-v2/revisions/5de55d3804cf/schema)