Transcription result received

Description

A conversation recording transcription result was received: automatic by rules or on a transcribe request.

Table 1. System characteristics

Class

Type

connection_set_asr

Properties

Table 2. Properties
Specification Description

Name: id
Type: varchar
Empty value: — 

Dialog identifier (idconn in Token API ccs requests) in <cid>.<invitets> format.

Name: dialog_text
Type: varchar
Empty value: — 

Conversation text by utterances of both sides.

Name: dialog_json
Type: json
Empty value: — 

Conversation utterances: list of side objects (a, b), b, e — start and end, sec; prev, next — the utterance overlaps with the previous or next utterance of the other side; text.

Name: json
Type: json
Empty value: — 

Phrases by side: {"a": [Phrase], "b": [Phrase]}, Phrase — b, e, text, words (list of b, e, word).

Name: txt_a
Type: varchar
Empty value: — 

Party A text.

Name: txt_b
Type: varchar
Empty value: — 

Party B text.

Example

The example is based on the code: the event was not received when capturing examples on the test bench.

[
  "notify",
  {
    "sid": "my_subscr",
    "class": "ccsevents",
    "type": "connection_set_asr",
    "eventts": 1792160450760,
    "data": {
      "id": "94f566bb-dc51-4fa2-a53b-efb7dd1e40fb.1792160517710",
      "dialog_text": "A: good afternoon\r\nB: hello",
      "dialog_json": [
        {
          "side": "a",
          "b": 0.4,
          "e": 1.2,
          "prev": false,
          "next": false,
          "text": "good afternoon"
        },
        {
          "side": "b",
          "b": 1.6,
          "e": 2.5,
          "prev": false,
          "next": false,
          "text": "hello"
        }
      ],
      "json": {
        "a": [
          {
            "b": 0.4,
            "e": 1.2,
            "text": "good afternoon",
            "words": [
              {
                "b": 0.4,
                "e": 0.8,
                "word": "good"
              },
              {
                "b": 0.8,
                "e": 1.2,
                "word": "afternoon"
              }
            ]
          }
        ],
        "b": [
          {
            "b": 1.6,
            "e": 2.5,
            "text": "hello",
            "words": [
              {
                "b": 1.6,
                "e": 2.5,
                "word": "hello"
              }
            ]
          }
        ]
      },
      "txt_a": "good afternoon",
      "txt_b": "hello"
    }
  }
]

See also