Additional metadata on requesting callback for large audio files

[pranjal-zipteams]

Hi, I'm using the Python SDK and transcribing very large type files (typically a few hundred MBs).

So far I have successfully:

• Retrieved the transcripts of the calls on a separate endpoint using the callback URI
• Made use of the request_id and file_name mapping created on the origin, to identify and remap transcripts with audio files.

What I want to know:

• Is there a way we can send additional custom metadata/parameters while requesting a callback to a specific URI?
• If so, could you please point me to the reference of what options to use for this?

Thanks!

[BrianHillis-DG]

Hey @pranjal-zipteams, it sounds like the tag feature is what you're looking for.

Python SDK Usage:

#single tag
response = await dg_client.transcription.prerecorded(source,
        { "model": "general", 
          "language": "en", 
          "tier": "enhanced",
          "tag" : "foo",
          "callback" : CALLBACK_URL
        })
        
#multiple tags
response = await dg_client.transcription.prerecorded(source,
        { "model": "general", 
          "language": "en", 
          "tier": "enhanced",
          "tag" : ["foo", "bar"],
          "callback" : CALLBACK_URL
        })

These tags appear on the callback in the metadata.tags field