Azure Cognitive Speech to Text Duplicate Sentences returned on Channel 0 and Channel 1

James Nicolson 5 Reputation points
2023-06-06T05:35:35.0933333+00:00

We are developing a solution using Azure Cognitive Speech to Text service and have an issue with duplicate sentences being returned. 

We have some cases with dual channel audio which appear to transcribe correctly with speaker channels. We have stereo audio in the input file.

I suspect the issue is the type of audio but is there any suggestions on why we may be seeing duplicate responses for a transcription for both channels?

The audio is a call recording between two people.

See example:

Example API request

{
    "properties": {
        "diarizationEnabled": false,
        "wordLevelTimestampsEnabled": false,
        "displayFormWordLevelTimestampsEnabled": false,
        "duration": null,
        "channels": null,
        "destinationContainerUrl": null,
        "punctuationMode": "DictatedAndAutomatic",
        "profanityFilterMode": "Masked",
        "timeToLive": "P0Y0M30DT0H0M0S",
        "diarization": null,
        "languageIdentification": null,
        "email": null,
        "error": null
    },
    "contentUrls": [
        "xxxxxxx"
    ],
    "locale": "en-GB",
    "displayName": "Transcription Data",
    "customProperties": {}

 Example API response transcription

{
  "source": "xxxxxxx",
  "timestamp": "2023-06-05T14:11:57Z",
  "durationInTicks": 12957400000,
  "duration": "PT21M35.74S",
  "combinedRecognizedPhrases": [
    {
      	"channel": 0,
      	"lexical": "hi victoria nice to see you today 
	}
	{ 	"channel": 1,       
		"lexical": "hi victoria nice to see you today
	}
Azure AI Speech
Azure AI Speech
An Azure service that integrates speech processing into apps and services.
1,762 questions
Azure AI Language
Azure AI Language
An Azure service that provides natural language capabilities including sentiment analysis, entity extraction, and automated question answering.
421 questions
{count} vote

1 answer

Sort by: Most helpful
  1. Radhika Jagtap 20 Reputation points
    2024-07-12T12:43:06.1633333+00:00

    I am also facing same issue. Please suggest.

    We are referring this github code: Batch ranscription approach - REST API

    https://github.com/Azure-Samples/cognitive-services-speech-sdk/tree/182fabecee81a4af948e6a049db84da60cb09181/samples/batch/csharp

    0 comments No comments

Your answer

Answers can be marked as Accepted Answers by the question author, which helps users to know the answer solved the author's problem.