Hello, we are experiencing an issue with VAPI. The assistant suddenly stops speaking even though the caller continues to speak. Call id: 04e9afd3-edc0-4e64-b7e9-20477f1cc018 jan 13., 17:29 I will write the configuration details below 👇

message has been deleted

What specific settings are needed?

@august To better assist you with the issue of the assistant stopping unexpectedly, I would need to know more about your current VAPI configuration settings. Specifically, details about the following would be helpful: 1. The version of VAPI you are using. 2. Any relevant timeout settings besides

VAPI_SILENCE_TIMEOUT_SECONDS

. 3. The specific conditions or triggers that seem to cause the assistant to stop speaking. 4. The type of content the assistant is handling when the issue occurs (e.g., streaming audio, pre-recorded messages, etc.). 5. Any error messages or logs that you might have observed when the issue occurs. These details will help me understand the context better and provide you with a more accurate solution. If you have any other information that you think might be relevant, please feel free to share that as well.

@august Looking at the screenshot, you can see there was no response from your side (custom-llm) after the user's last transcript. https://cdn.discordapp.com/attachments/1329748561088348243/1330017632258359439/Screenshot_2025-01-18_at_09.03.35.png?ex=678c72f2&is=678b2172&hm=f19d78b0b5579ac960fa55bce99f0108d4b7aa54810e211608f14319af1edd46& https://cdn.discordapp.com/attachments/1329748561088348243/1330017632497438740/message.txt?ex=678c72f2&is=678b2172&hm=6735345e4c4a650da1a60a987f0296b50b175e589e80bc957923d32f78ae26b9&

@Shubham Bajaj Is it UTC time (13:32), right?

@august I looked into the logs again, and they remain unchanged. I don't have examples since the server did not respond. Could you try another call, logging API request-response and check your server response as well to maintain uniformity.

@Shubham Bajaj We interact with VAPI on a daily basis. We have numerous calls, and this seems to be a one-off case. We’ve thoroughly checked the logs of Nginx, load balancer, and the logs of application, but there are no errors on our side. And, as far as I know, such issues haven’t occurred before. That’s why I asked if you have an example of this kind of error so we could try to find something in common. But again, so far, it seems like the problem is on the VAPI side.

@august People are using custom-LLM on a daily basis, and I understand it might lead to frustration if you can't find a solution. Honestly speaking, we just handle the request-response cycle, but if this issue appears again for you or if I see it somewhere else, I will update you. In case you want to have a 1:1 discussion, we can connect today during Discord office hours.

@Shubham Bajaj Alright, I understand, thanks. I’ll be waiting for updates. I’m here and available if anything comes up.

Please also respond to your other 5xx/4xx status code related ticket if you haven't done so already.