Hello Vapi Support, We have a recurrent issue on PSTN inbound calls where the caller is clearly speaking (audible in the call recording waveform), but Vapi does not create a new user turn and the transcriber shows no transcripts/events during that segment. The assistant stays silent for a long time (e.g., 50+ seconds). Call ID: 019c143f-4c46-7ee5-aac5-f12d774ff3d3 Timestamp: around minute ~3:00 the caller repeatedly says “hola/hello”, audible in the recording waveform, but there are no transcriber/endpointing user-turn events. Additional details: In the dashboard, downloading “PCAP” only provides a sip.pcap (SIP signaling only). We need a full PCAP including RTP to verify inbound audio and packet stats (jitter/loss) during the failure window. Logs show voice (ElevenLabs) websocket activity, but no transcriber activity during the caller speech segment. Could you please: Confirm whether inbound RTP audio was received by Vapi during the failure window (and provide jitter/loss/packet stats if available). Explain why the VAD/endpointing did not detect speech / create a user turn even though the caller is audible. Provide (or enable) a downloadable PCAP including RTP for this call (not SIP-only). Thanks.

To diagnose why VAD/endpointing did not detect speech for your call, start by retrieving the full call configuration and event logs using the Get Call API endpoint. This will show the exact VAD/transcriber settings and any detected events for your call:

bash
curl -X GET "https://api.vapi.ai/v1/calls/019c143f-4c46-7ee5-aac5-f12d774ff3d3" \
  -H "Authorization: Bearer YOUR_API_KEY"

Review the returned JSON for

vadAssistedEndpointingEnabled

,

confidenceThreshold

, and transcriber events to identify any misconfigurations or missing detections. This is the most direct way to begin troubleshooting the issue based on your call's actual pipeline configuration and logs. For more, see the [Get Call API documentation](https://docs.vapi.ai/api-reference/calls/get-call). Source: - [Get Call API documentation](https://docs.vapi.ai/api-reference/calls/get-call)

Please we need support

Hey, can you please try the DeepGram Nova 2 or Nova 3 transcriber? I can see you are using OpenAI transcriber. Deepgram offers the best consistency and supports most languages. I have checked, although there were some minor packet losses, the audio was still clear, and it's an issue with the transcriber.

Hey, I am having packet loss issue too, can you please check this https://discord.com/channels/1211482211119796234/1468504819110252584

Hey Pranav, can you send the call URLs and the affiliated call ID to our support mail, as they can be private- support@vapi.ai , add sarthak.vaish@vapi.ai in cc.

Hey Sarthak, I have cced you in email with Subject :

VAPI Packet Loss on Daily SFU issue (identified cause) needs resolution

can you please check this as early as possible, I am very sure this could create big p0 for web call integeration folks. it is an infra problem and there has to be stability to it, we have our deals blocked on this one factor that there is too much glitch. I have written detailed RCA in the email

Alright, assigned myself to that thread, will investigate and get back to you!

Great, Thank you. Will wait for the update