Empty transcript array in /api/v2/analytics/conversations/transcript endpoint

I’m hitting a wall trying to pull the full voice-to-text transcript for a specific conversation ID via the Speech and Text Analytics API. I have the conversationId from our adherence logs, and I can see the analytics summary just fine. The problem is getting the actual text content.

Here is the endpoint I’m calling:

GET /api/v2/analytics/conversations/conversations/{conversationId}/transcript

I’m getting a 200 OK response, but the transcript array inside the JSON payload is completely empty. The text field is null for every segment.

{
 "transcript": [
 {
 "start": "2023-10-27T14:30:00.000Z",
 "end": "2023-10-27T14:30:05.000Z",
 "text": null,
 "speaker": "customer"
 }
 ]
}

I know the conversation was transcribed because I can see it in the UI. I’ve checked the API docs and I’m using the correct application/json header. Is there a delay for the transcript to populate in the API even after the conversation is closed? Or am I missing a query parameter to force it to return the text?

Also, I’m working Central time, so maybe there’s a timezone issue with the date range? I tried adding from and to params but that didn’t help.