Call Imports
Flip the user <-> agent mapping on a diarised row.
The worker's "first speaker is the agent" heuristic is right most of
the time but does fail on inbound recordings where the customer
greets first, on recordings where the agent stays silent for the
intro, etc. Rather than rerun the (expensive) STT + pyannote
pipeline for those cases, we let reviewers flip the mapping in
place: the structured diarised_segments are the source of truth
and we re-render the plain-text diarised_transcript from them
with the swap applied. The next CSV export will then show the
corrected labels.
Returns the updated row so the frontend can refresh without an extra round-trip.
AuthorizationBearer <token>
JWT access token. Use: Authorization: Bearer .
In: header
Path Parameters
call_import_id*Call Import Id
Format
uuidrow_id*Row Id
Format
uuidHeader Parameters
X-API-Key?string|null
X-EFFICIENTAI-API-KEY?string|null
X-Workspace-Id?string|null
Response Body
application/json
application/json
curl -X POST "https://example.com/api/v1/call-imports/497f6eca-6276-4993-bfeb-53cbbbba6f08/rows/497f6eca-6276-4993-bfeb-53cbbbba6f08/diarised-speaker-swap"{ "id": "497f6eca-6276-4993-bfeb-53cbbbba6f08", "row_index": 0, "conversation_id": "string", "recording_url": "string", "recording_date": "2019-08-24", "transcript": "string", "transcript_source": "string", "transcript_provider": "string", "transcript_model": "string", "transcript_status": "string", "transcript_error": "string", "transcribed_at": "2019-08-24T14:15:22Z", "diarised_transcript": "string", "diarised_transcript_provider": "string", "diarised_transcript_model": "string", "diarised_transcript_status": "string", "diarised_transcript_error": "string", "diarised_at": "2019-08-24T14:15:22Z", "diarised_llm_provider": "string", "diarised_llm_model": "string", "diarised_prompt": "string", "diarised_segments": [ {} ], "diarised_speaker_swap": false, "status": "pending", "recording_s3_key": "string", "recording_content_type": "string", "recording_size_bytes": 0, "error_message": "string", "attempts": 0, "raw_columns": {}, "created_at": "2019-08-24T14:15:22Z", "updated_at": "2019-08-24T14:15:22Z"}Community & contact
- Found a bug or have a feature request? Open a GitHub issue.
- Join our Discord for faster replies!