Skip to main content

Scoring

Scoring returns a numeric QA result plus item-level reasons and evidence. Use it for review queues, coaching, and quality reporting.

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/ai-score \
  -H 'Authorization: Bearer $MINIVOICE_API_KEY'

Real Response​

{
  "data": {
    "status": "completed",
    "total_score": 17,
    "max_score": 20,
    "percentage": 85,
    "items": [
      {
        "name": "Greeting",
        "score": 5,
        "max_score": 5,
        "reason": "Agent greeted the caller clearly.",
        "evidence": "Thanks for calling support."
      }
    ]
  }
}

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​

Create a QA dashboard where supervisors filter low-scoring calls, inspect item evidence, and compare the score with the transcript.

Common Failure Case​

{
  "error": {
    "code": "ai_score_not_found",
    "message": "AI score 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

Show the overall percentage for scanning, but never hide item-level reasons and evidence. Reviewers need the evidence text to decide whether the score is useful, especially when a call is short, noisy, or only partially resolved.

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.