Skip to main content

Transcription

Transcription turns recorded call audio into searchable text. It is the foundation for summaries, scoring, review workflows, and support search.

AI results are read after a call has completed and recorded audio has been processed. The public API exposes read endpoints for combined AI, transcription, summaries, scoring, and topics. Your integration should store the returned status and payload so the UI can show completed, missing, and failed states separately.

Real Request​

curl -sS https://api.minivoice.eu/v1/calls/call_123/transcription \
  -H 'Authorization: Bearer $MINIVOICE_API_KEY'

Real Response​

{
  "data": {
    "status": "completed",
    "language": "en",
    "full_text": "Hello, I need help understanding my invoice.",
    "segments": [
      {
        "speaker": "caller",
        "text": "Hello, I need help understanding my invoice.",
        "start_ms": 0,
        "end_ms": 3200
      }
    ]
  }
}

Real Webhook Example​

{
  "event": "call.completed",
  "created_at": "2026-05-29T12:10:00Z",
  "call": {
    "id": "call_123",
    "customer_id": "cust_123",
    "application_id": "app_123",
    "status": "completed",
    "direction": "outbound",
    "from": "+15551230001",
    "to": "+15551230002"
  },
  "variables": {},
  "data": {}
}

Common Use Case​

Store full_text with the call ID so support agents can search prior conversations and open the original call when context is needed.

Common Failure Case​

{
  "error": {
    "code": "transcription_not_found",
    "message": "transcription not found"
  }
}

Do not hide the call when one AI result is missing. Show the call metadata immediately and let each AI section render its own completed or unavailable state.

Response handling

AI endpoints should be consumed as enrichment, not as the primary call record. Store the call ID, result status, and raw response for each feature your product uses. A completed call can have a completed transcript and a missing score, or a completed summary and no topics. Model those states independently in your UI and data layer.

For dashboards, prefer progressive loading. Show the call metadata first, then render transcript, summary, score, sentiment, and topics as each read succeeds. For automations, branch on explicit result values and status fields instead of assuming every recorded call has the same AI outputs.

Display guidance

For long calls, show transcript text in a scrollable call detail area and keep the call metadata visible nearby. If segments are available, use them for timestamps or speaker turns, but keep full_text searchable for simple integrations.

Testing​

Place a recorded call with known wording, wait for call.completed, then fetch this endpoint. Save the response as a fixture and test your parser against completed data plus the not-found response shown above.