Analyzer deploys successfully but silently ignores key schema fields - Azure AI Content Understanding (API version 2025-05-01-preview) in the West US region

Hi Dennis

Good day. I am trying to replicate above issue

It seems that some properties in your schema for pre-built audio analyzer are not working.

Ask

1. I think you are using rest API to send the payload to analyzer endpoints. Could you share full rest api (Curl command /SDK )
2. Reference for below properties

  "properties": {
        "mode": "verbatim",
        "includeFillerWords": true,
        "includeSpeakerDiarization": true,
        "includeWordTimestamps": true,
        "enablePunctuation": true
    }

Looking forward to hearing from you.

Thank you

Core code is from here: https://github.com/Azure-Samples/azure-ai-content-understanding-python

Here's my code to deploy the analyzer:

from python.content_understanding_client import AzureContentUnderstandingClient

credential = DefaultAzureCredential()

client = AzureContentUnderstandingClient(
    endpoint         = AZURE_AI_ENDPOINT,
    api_version      = AZURE_AI_API_VERSION,
    subscription_key = AZURE_AI_API_KEY,
    x_ms_useragent = "azure-ai-content-understanding-python/content_extraction"
)

verbatim_tx_json_filename = "verbatim_tx.json" # Analyzer schema in 1st post above
with open(verbatim_tx_json_filename, "r") as f:
    verbatim_tx = json.load(f)
print(f"Serialized JSON Analyzer schema from file - {verbatim_tx_json_filename}:\n",
        json.dumps(verbatim_tx, indent = 3))

analyzer_id = "verbatim-tx-" + str(uuid.uuid4()) # generate a unique analyzer ID
try: # create the custom analyzer
    print(f"\n🔨 Creating custom analyzer: {analyzer_id}")
    
    response = client.begin_create_analyzer(analyzer_id, verbatim_tx)
    result   = client.poll_result(response)
    
    print("✅ Analyzer created & deployed successfully:")
    print(json.dumps(response.json(), indent = 3))
    
except Exception as e:
    print(f"\n❌ Error creating analyzer: {e}")
    analyzer_id = None         # set to None if creation failed

response_data = response.json()
print(response_data["mode"].upper(), "MODE")

Hi Dennis

Could you also quota the reference used for below properties

"properties": { "mode": "verbatim", "includeFillerWords": true, "includeSpeakerDiarization": true, "includeWordTimestamps": true, "enablePunctuation": true }

Thank you

Hi Dennis

Below is sample schema for audio transcription

Reference - https://github.com/Azure-Samples/azure-ai-content-understanding-python/blob/main/analyzer_templates/audio_transcription.json

Please share your reference used properties as mentioned in above comment.

{
  "description": "Sample call transcript analyzer",
  "scenario": "callCenter",
  "config": {
    "returnDetails": true,
    "locales": ["en-US"]
  },
  "fieldSchema": {
    "fields": {
      "Summary": {
        "type": "string",
        "method": "generate"
      },
      "Sentiment": {
        "type": "string",
        "method": "classify",
        "enum": [ "Positive", "Neutral", "Negative" ]
      },
      "People": {
        "type": "array",
        "description": "List of people mentioned",
        "items": {
          "type": "object",
          "properties": {
            "Name": { "type": "string" },
            "Role": { "type": "string" }
          }
        }
      }
    }
  }
}

API version 2025-05-01-preview docs are preliminary and kind of sketchy as expected. Reviewed most of the samples/examples on GitHub. The SDK hasn't been publicly released, per my searching. Querying GitHub Copilot for Azure Content Understanding "prebuilt-audioAnalyzer" properties documentation returns: