https://vapi.ai/ logo
Join Discord
Powered by
Web Call Error Meaning
# support
e
I am not seeing any documentation on the meaning of 
call.in-progress.error-assistant-did-not-receive-customer-audio
. What is the specific meaning of this error? I never triggered this in development and I am now getting a fair number of these in production. Based on the message it seems these calls were started successfully, but I assume this error is different than a silence timeout and/or denying microphone permission. Example of calls where this is occurring: 
cb2372c9-ad27-4526-8cc0-a5a030b4f2bc
eddd6e2a-c3f7-40c8-951f-f183eac42ee3
017aa81a-f5c4-49f6-adc0-b7d83b16d1c9
7ae64ee7-70f7-48db-97a2-8e7f9193d1dc
fb26b047-0847-4083-85de-29ba22018245
v
Hi there! 👋 I understand you're seeing the 
call.in-progress.error-assistant-did-not-receive-customer-audio
error in production and need clarification. While I can't analyze your specific call IDs, I can definitely explain what this error means and how it differs from other similar issues. ## What this error means The 
call.in-progress.error-assistant-did-not-receive-customer-audio
error indicates that the AI assistant did not receive audio from the customer during the call, even though the call was successfully established. This is a technical failure in the audio transmission process rather than a timeout or permission issue. According to the information I found, this error can occur due to several reasons: - A missing server URL configuration for the assistant to receive audio data from the customer - Network errors or issues with the websocket connections - Problems with the Deepgram transcription service (possibly blocking your IP address) - Incorrect configuration of the assistant ID on the phone number ## How it differs from silence timeout The key difference is that a silence timeout is a deliberate action based on a timer, while this error indicates a technical failure: - **Silence timeout**: Occurs when the assistant doesn't detect any speech from the user for a specified duration (
silenceTimeoutSeconds
, default 30 seconds). This is a graceful fallback when users don't respond, and logs will show "Silence Timed Out. Ending Call..." - **Assistant did not receive audio error**: Indicates the assistant isn't receiving the audio stream at all due to technical issues, even if the user is speaking. The audio transmission path is broken. ## How it differs from microphone permission denial - **Microphone permission denial**: Results in a specific error 
customer-did-not-give-microphone-permission
and occurs when the user explicitly denies microphone access. This typically happens within 10 seconds if permission isn't granted.
- **Assistant did not receive audio error**: Can occur even when microphone permissions are granted but the audio isn't reaching the assistant due to network issues, configuration problems, or other technical failures. ## Troubleshooting steps Here are some steps you can take to resolve this issue: 1. **Check assistant ID configuration**: Ensure the assistant ID is correctly configured and assigned to the phone number. 2. **Investigate network issues**: Check for network or service issues that might be affecting audio transmission. 3. **Review server URL configuration**: Make sure you have a properly configured server URL for the assistant to receive audio data. 4. **Check for Deepgram issues**: If you're using Deepgram, it might be blocking your IP address. Try using Vapi's key instead. 5. **Adjust configuration settings**: Consider modifying your assistant configuration:
Copy code
json
   "start_speaking_plan": {
     "waitSeconds": 0.8,
     "smartEndpointingEnabled": true
   }
Copy code
json
   "transcriber": {
     "model": "nova-2",
     "language": "en",
     "provider": "deepgram",
     "endpointing": 300
   }
6. **Enable recording**: Enable recording to help troubleshoot the issue by setting 
recordingEnabled
to true. Is there anything specific about these troubleshooting steps you'd like me to elaborate on? Sources: - [Discord Thread about Assistant Not Receiving Customer Audio](https://discord.com/channels/1211482211119796234/1211483291191083018/threads/1296435541952303124) - [Discord Thread about Potential Network or Deepgram Issues](https://discord.com/channels/1211482211119796234/1211483291191083018/threads/1340235729204285510) - [Discord Thread about Assistant Configuration Problems](https://discord.com/channels/1211482211119796234/1211483291191083018/threads/1336652401733668864) - [Discord Thread about Silence Timeout](https://discord.com/channels/1211482211119796234/1211483291191083018/threads/1296852804631265392)
e
These are web calls the issues are occurring with, not phone calls
v
checking if this is resolved/solved for you?
e
@Shubham Bajaj Still occurring, just wondering exactly how this status is triggered
How this is different from a silence timeout, etc
I see the automated response here, wanted to confirm this was correct though as I couldn't find any documentation on it
v
🔵 23:43:21:578 Live Call Manager Ending...([call.in](http://call.in)-progress.error-assistant-did-not-receive-customer-audio) You're seeing an error message that indicates a web call is ending prematurely because the system didn't receive any audio from the customer's microphone within the expected time window. This issue is typically caused by either microphone permission problems, connectivity issues, or configuration settings with the timeout being too short. We'll need to check the user's browser permissions, network connectivity, and potentially adjust the timeout settings if needed.
If this is happening consistently, you may need to increase the customerJoinTimeoutSeconds parameter for your assistant.
Copy code
// When configuring your assistant
   const assistantConfig = {
     // ... other config
     customerJoinTimeoutSeconds: 30, // Increase from default 15 seconds
     // ... other config
   };
Hey Egreet, could you make the customer join timeout second chance? Let me know how it goes for you.
e
@Shubham Bajaj Awesome thank you!
v
Marking this ticket as Solved ✅
2 Views
PreviousNext
Powered by