I am having an issue with transcribing (Speech-To-Text) an audio file hosted on Azure Storage container from the Cognitive Services API.
The services are of the same resource (and I created a VNet and they are part of the same subnet).
After I take the response from there the contentUrl:
The error I get is:
{
"successfulTranscriptionsCount": 0,
"failedTranscriptionsCount": 1,
"details": [
{
"source":"https://{service-name}.blob.core.windows.net/meetingnotes/Meeting82035.wav",
"status": "Failed",
"errorMessage": "Error when downloading the recording URI. StatusCode: Conflict.",
"errorKind": "DownloadRecordingsUrisUnknownError"
}
]
}
I contacted Azure support and they provided the correct solution, which is to add the Role “Storage Blob Data Contributor” to the speech services resources.
Go to IAM of your storage account
Go to Role Assignments
click "Add", then add your speech service in Managed Identities. That should fix it.