No Analytics available in cognitive-services-speech-sdk

[rakeshdevineni123]

IN ORDER TO ASSIST YOU, PLEASE PROVIDE THE FOLLOWING:

• Speech SDK log taken from a run that exhibits the reported issue.
  See instructions on how to take logs.

• A stripped down, simplified version of your source code that exhibits the issue. Or, preferably, try to reproduce the problem with one of the public samples in this repository (or a minimally modified version of it), and share the code.

• If relevant, a WAV file of your input audio.

• Additional information as shown below

Describe the bug

I have this deployment which I created by following the above steps. The output json does not have the Text analytics even though I've specified the Service keys and the endpoint.

https://portal.azure.com/#view/HubsExtension/DeploymentDetailsBlade/~/overview/id/%2Fsubscriptions%2Fa61839b4-05a1-4aa3-9a3b-03c40b076a23%2FresourceGroups%2FTRA-DEV-ML-WORKBENCH%2Fproviders%2FMicrosoft.Resources%2Fdeployments%2FMicrosoft.Template-20241213103421/packageId/Microsoft.Template/packageIconUri/https%3A%2F%2Fcatalogartifact.azureedge.net%2Fpublicartifacts%2FMicrosoft.Template-2.1.8%2FMedium.png/primaryResourceId//createBlade~/%7B%22bladeName%22%3A%22CustomDeploymentBlade%22%2C%22extension%22%3A%22Microsoft_Azure_CreateUIDef%22%2C%22telemetryId%22%3A%22314f2b48-c377-4d68-9f2c-e0cbaace4409%22%7D

To Reproduce

Steps to reproduce the behavior:

1. /~https://github.com/Azure-Samples/cognitive-services-speech-sdk/blob/master/samples/ingestion/ingestion-client/infra/guide.md

Expected behavior

I expected the Json output to also have this component as we subscribe to that service. But it is not present.
"conversationAnalyticsResults": {
"conversationSummaryResults": {
"conversations": [

Version of the Cognitive Services Speech SDK

Which version of the SDK are you using.

Platform, Operating System, and Programming Language

• OS: [e.g. Windows, Linux, Android, iOS, ...] - please be specific
• Hardware - x64, x86, ARM, ...
• Programming language: C#, C++, Java, JavaScript, Objective-C, Python
• Browser [e.g. Chrome, Safari] (if applicable) - please be specific

Additional context

• Error messages, stack trace, ...
• Any additional information.

[HenryvanderVegte]

Hi @rakeshdevineni123 ,

could you provide the values you set for the following parameters of the template:

• Locale
• Custom Model Id (no need to provide the full id, but for repro it'd be helpful to know if a custom model was used)
• Profanity Filter Mode
• Punctuation Mode
• Add Diarization
• Add Word Level Timestamps
• Sentiment Analysis
• Pii Redaction

Thanks!

[rakeshdevineni123]

Locale - en-US | English (United States)
Custom Model Id (no need to provide the full id, but for repro it'd be helpful to know if a custom model was used) - left blank
Profanity Filter Mode - Masked
Punctuation Mode - None
Add Diarization - true
Add Word Level Timestamps - false
Sentiment Analysis - AudioLevel
Pii Redaction - UtteranceAndAudioLevel

[HenryvanderVegte]

Hi @rakeshdevineni123 ,

I tried to repro with the settings you described, and I can receive sentiment results as well as redacted text, see 'sentiment' below:

{
  "source": "***",
  "timeStamp": "2025-01-14T10:46:06Z",
  "durationInTicks": 74800000,
  "duration": "PT7.48S",
  "combinedRecognizedPhrases": [
    {
      "channel": 0,
      "lexical": "",
      "itn": "",
      "maskedITN": "",
      "display": "*******",
      "sentiment": {
        "negative": 0.0,
        "neutral": 1.0,
        "positive": 0.0
      }
    }
  ]
...

Based on your description here

I expected the Json output to also have this component as we subscribe to that service. But it is not present.
"conversationAnalyticsResults": {
"conversationSummaryResults": {
"conversations": [

This requires you to enable the conversational text analytics features, which are not exposed via parameters.
I see that there are 4 conversational text analytics related properties in the ARM template ("ConversationPiiCategories", "ConversationPiiRedaction","ConversationPiiInferenceSource", "ConversationSummarizationOptions"):

/~https://github.com/Azure-Samples/cognitive-services-speech-sdk/blob/d5f0d936fd2065e00499ba8725500738a3b1cce2/samples/ingestion/ingestion-client/infra/main.json#L261C70-L268C347

Can you try setting those manually in your ARM template?

Thanks!

[github-actions]

This item has been open without activity for 19 days. Provide a comment on status and remove "update needed" label.