Streaming stuck on pending state.

[bhupenchn]

We are facing this issue intermittently, and this has increased in the last week.

Programming Language : Python, Deepgram SDK v3.10.1
Request IDs :

• 8a7127c6-2943-4f33-84dc-4dbb04087620
• 3a181d0e-8232-46a2-8538-7872c1a71cc0
• f835cc19-bc04-443b-9c19-32c987c49d9e

This is for the streaming STT endpoint.

The states have now changed to Unknown, can you help me understand why do these happen? And the same codebase works in most cases but sometimes these issues emerge.

From similar search discussions previously I see

The pending state can mean a few things. (a) It could mean the request has completed and we're still processing your logs, (b) it could mean that the request is in progress (this is often the case for voice agent requests), or (c) in this case it means that you sent empty audio and the websocket closed unexpectedly. I would double check your configuration and make sure you're sending audio properly.

(a), (b) seem unlikely, since it has now changed to Unknown state
(c) This seems unlikely as well, since we see in our logs that we have received and interim result, but there was no final message for one of the requests.

[deepgram-community[bot]]

Thanks for asking your question. Please be sure to reply with as much detail as possible so the community can assist you efficiently.
_{Consider joining our Discord community for more opportunity to engage with your fellow Deepgram users. You can earn points which can be redeemed for cool stuff by being active in our communities!}

[deepgram-community[bot]]

Hey there! It looks like you haven't connected your GitHub account to your Deepgram account. You can do this at https://community.deepgram.com - being verified through this process will allow our team to help you in a much more streamlined fashion.

[deepgram-community[bot]]

It looks like we're missing some important information to help debug your issue. Would you mind providing us with the following details in a reply?

• The deepgram product you are using (e.g Speech to Text, Agent API)

[bhupenchn]

It is Speech to Text

[jkroll-deepgram]

Hi @bhupenchn, this typically occurs when a streaming connection is not closed neatly. It may also happen when you open a connection, do not send any audio, and then the connection later fails.

Are you getting any transcription responses from Deepgram for these streams?

Can you share any request IDs where this occurred?

[bhupenchn]

Thanks for the response.

We did see interim results for a few of the requests. Some are here.

8a7127c6-2943-4f33-84dc-4dbb04087620
3a181d0e-8232-46a2-8538-7872c1a71cc0
f835cc19-bc04-443b-9c19-32c987c49d9e

[bhupenchn]

Any updates on this?

[jkroll-deepgram]

Hi @bhupenchn, thanks for your patience. I can confirm that Deepgram is receiving empty or non-audio data from you. I'm seeing this metadata on your audio:

    "description": "empty",
    "mimetype": "inode/x-empty",

Please troubleshoot how you are sending audio to Deepgram, and the contents of the audio you're intending to send.

[bhupenchn]

Hmm, but our pipeline remains same across all calls, this issue is persistent on some requests but not on all, you believe that this issue is because of the empty data?

[jkroll-deepgram]

@bhupenchn yes it may be. One example I've seen is from customers who have web applications, so different end users have different operating systems, browsers, microphones, and security configurations. That causes nearly all properties of the audio to vary, and not be entirely within the application's control.