Purpose

Use this endpoint to retrieve Speech Analytics for a specific CDR, including:

Summary
Score
Transcription (utterances with timestamps and participant type)

This endpoint also supports multiple output formats and optional translation when available.

Scope: reporting.cdrs.read

Prerequisites

The CDR ID. Use Get list of CDRs to retrieve CDR IDs.
Authentication using a Bearer token. See: Base URL and Authentication.

Notes

An utterance is one continuous segment of speech from a single participant, returned with start and end timestamps.
In most cases, utterances correspond to speaker turns, but segmentation can vary depending on the transcription model.

Endpoint

GET /api/v4/reporting/cdrs/{cdr_id}/speech_analytics

Path parameters

cdr_id The unique identifier of the CDR (UUID).

Query parameters

Optional:

language

Language code for translation, for example en or es. If not specified, the response returns the original language. Must be one of the languages supported by your contact center (callcenter).

Output formats

This endpoint supports these response formats via the Accept header:

application/json
text/vtt
application/x-subrip
text/plain

Example requests

JSON

curl -X GET "https://{cluster_id}.voiso.com/api/v4/reporting/cdrs/f0ed7183-cddb-46c5-9037-19e16126764f/speech_analytics" \
  -H "Authorization: Bearer <contact_center_api_key>" \
  -H "Accept: application/json"

JSON with translation

curl -X GET "https://{cluster_id}.voiso.com/api/v4/reporting/cdrs/f0ed7183-cddb-46c5-9037-19e16126764f/speech_analytics?language=en" \
  -H "Authorization: Bearer <contact_center_api_key>" \
  -H "Accept: application/json"

WebVTT

curl -X GET "https://{cluster_id}.voiso.com/api/v4/reporting/cdrs/f0ed7183-cddb-46c5-9037-19e16126764f/speech_analytics" \
  -H "Authorization: Bearer <contact_center_api_key>" \
  -H "Accept: text/vtt"

SRT

curl -X GET "https://{cluster_id}.voiso.com/api/v4/reporting/cdrs/f0ed7183-cddb-46c5-9037-19e16126764f/speech_analytics" \
  -H "Authorization: Bearer <contact_center_api_key>" \
  -H "Accept: application/x-subrip"

Plain text

curl -X GET "https://{cluster_id}.voiso.com/api/v4/reporting/cdrs/f0ed7183-cddb-46c5-9037-19e16126764f/speech_analytics" \
  -H "Authorization: Bearer <contact_center_api_key>" \
  -H "Accept: text/plain"

Response

JSON response example

{
  "speech_analytics": {
    "id": "f0ed7183-cddb-46c5-9037-19e16126764f",
    "summary": {
      "language": "es",
      "text": "El cliente llamó por error, el agente se presentó como representante de la empresa Example, y el cliente se disculpó."
    },
    "score": {
      "value": 2
    },
    "transcription": {
      "language": "es",
      "utterances": [
        {
          "start_time": "2026-01-20T18:19:45.202Z",
          "end_time": "2026-01-20T18:19:45.522Z",
          "text": "Hola!",
          "participant_type": "contact"
        }
      ]
    }
  }
}

Example outputs by format

The same Speech Analytics transcription can be returned in different formats depending on the Accept header.

WebVTT (Accept: text/vtt)

WEBVTT

1
00:00:00.000 --> 00:00:00.320
[contact] Hola!

SRT (Accept: application/x-subrip)

1
00:00:00,000 --> 00:00:00,320
[contact] Hola!

Plain text (Accept: text/plain)

2026-01-20T18:19:45.202Z to 2026-01-20T18:19:45.522Z contact: Hola!

Troubleshooting

406 Not acceptable

Returned when the requested output format is not supported. The error message indicates the supported formats.

409 Conflict

Returned when language was requested but the translation is not available yet.

404 Not found

Returned when the CDR does not exist or is not accessible.