Problem:
The main problem is identifying the right Microsoft Azure components and then integrating them into the current Agora Speech-to-Text start request payload. The goal is to successfully upload a .vtt file into the MS Azure container after every STT stop request.
Solution:
Step 1: Create a storage account using MS Azure dashboard and save as your accessKey.
Step 2: Click the newly created storage account and navigate into its main menu and select Security + networking tab, then Access keys. Save the key1 or key2 string values as your secretKey
Step 3: Also in the storage account main menu, please scroll down, open Data storage and select "Containers"
Step 4: Create a new "container" and save as your bucket name
Step 5: Use the saved credentials on the STT start request || captionConfig:
{
"captionConfig": {
"storage": {
"accessKey": "{{AccessKey(Azure)}}",
"secretKey": "{{SecretKey(Azure)}}",
"bucket": "{{Bucket(Azure)}}",
"vendor": 5,
"region": 0,
"fileNamePrefix": [
"STT",
"{{channelName}}"
]
}
}
}
Step 6: perform a full STT request lifecycle: acquire, start, query and stop then check if the .vtt files are uploaded into your bucket.