Get a Call Transcript
Use the get call transcript request to get the transcript for a specific captured call in your U-Capture system (speech to text).
Endpoint
Method: GET
URL: /transcript/post_call/{conversationID}
Path Parameter
Parameter Name | Data Type | Description | Required/Optional |
---|---|---|---|
| String | The ID of the conversation that you want to get the transcription for. | Required |
Response and Error Code
Response Code | Condition |
---|---|
200 | OK |
Response Parameter
For an example response see, Sample Response.
Parameter Name | Data Type | Description |
---|---|---|
| String | The ID of the conversation that you want to get the transcription for. |
| String | The ID of your tenant. |
| Array | The transcript for the conversation, see transcript Sub-parameters for further details. |
transcript
Sub-parametersParameter Name | Data Type | Description |
---|---|---|
| String | When the transcript started/when the speech started, in the format |
| Array | A list of the turns in the conversation, reflects the turns taken to speak by each participant, see turns Sub-parameters for further details. |
| String | The language of the speech in the transcript. |
turns
Sub-parametersParameter Name | Data Type | Description |
---|---|---|
| Integer | Which turn in the conversation, starting at |
| String | The type of participant speaking, for example |
| Integer | An offset to when conversation capture began, this indicates when the turn started/began to be spoken (in milliseconds). |
| Integer | An offset to when conversation capture began, this indicates when the turn ended/stopped being spoken (in milliseconds). |
| Array | A list of the transcribed words for the turn, see words Sub-parameters for further details. |
words
Sub-paramtersParameter Name | Data Type | Description |
---|---|---|
| String | The word transcribed from the conversation. |
| Array | An offset to when conversation capture began, this indicates when the word started/began to be spoken (in milliseconds). |
| String | An offset to when conversation capture began, this indicates when the word ended/stopped being spoken (in milliseconds). |
| Number | How confident the system is in the transcription for this word, |
Sample Response
Note
The below example response has been edited for documentation formatting purposes, in which the turns
array has been drastically shortened to only show two turns
consisting of only two words
each, typically transcripts will be longer.
{ "conversationId": "ca73aefb-389b-441d-8936-50bf45d7b88a", "tenantId": "66dd8418-b1d1-4d47-9218-c0e29bf13170", "transcript": { "timestamp": "2024-08-06T14:12:19.836Z", "turns": [ { "order": 1, "participantType": "Customer", "startOffset": 720, "endOffset": 1620, "words": [ { "text": "because", "startOffset": 720, "endOffset": 1200, "confidence": 0.95 }, { "text": "then", "startOffset": 1320, "endOffset": 1620, "confidence": 0.64 } ] }, { "order": 2, "participantType": "Agent", "startOffset": 2160, "endOffset": 2910, "words": [ { "text": "for", "startOffset": 2160, "endOffset": 2310, "confidence": 0.9 }, { "text": "holding", "startOffset": 2340, "endOffset": 2910, "confidence": 0.97 } ] } ], "lang": "en-us" } }