despite firstMessageMode: "assistant-speaks-first" Added firstMessageMode: "assistant-speaks-first" to assistantOverrides as suggested, but still complete silence from Vapi Speaker peer. Current state: Room opens, 3 peers present, stays connected 60 seconds My mic works (shows activity) Vapi Speaker peer present but never transmits audio Same assistant speaks perfectly when tested in Vapi dashboard Tested on multiple devices/browsers Payload: json{ "assistantId": "e66c1ef5-2f56-4530-a17e-8e644a7aa7f7", "assistantOverrides": { "firstMessageMode": "assistant-speaks-first", "customerJoinTimeoutSeconds": 60, "endCallFunctionEnabled": false, "silenceTimeoutSeconds": 600 } } Call ID: 17594425944850.677731884385207 This appears to be a backend provisioning issue. Can someone from Vapi engineering investigate why audio isn't routing to the Daily room despite correct configuration? Your code is correct. This needs backend investigation from Vapi's team.

Action Required: Since your configuration and code are correct but audio is still not routing, please report this to Vapi support with your Call ID (

17594425944850.677731884385207

), a brief description of the issue, and a screen recording if possible. This will help the engineering team investigate backend provisioning or routing problems. You can submit your report via the [Vapi issue reporting page](https://docs.vapi.ai/issue-reporting) for fastest resolution. Source: - [Vapi Issue Reporting Documentation](https://docs.vapi.ai/issue-reporting)

Message marked as helpful by @GIDEON! 🎉

Are you using a free vapi number?

Hi Jonathan, If you encounter any issues related to audio or voice settings in our platform, there are several recent updates that might be helpful. Here are a few options you should consider: 1. **Custom Audio in Call Transfers**: You can specify a custom audio file using the

transferCompleteAudioUrl

in

TransferPlan

when using the

warm-transfer-experimental

mode. This can be useful if you need to add custom notifications after transferring a call. More details [here](https://docs.vapi.ai/changelog/2025/5/25). 2. **Voice Model Customization**: If you're having trouble with voice quality or emotions, consider using the Cartesia voice experimental controls to adjust speed and emotional range. Learn more about this setup [here](https://docs.vapi.ai/changelog/2025/2/25). 3. **Improved Transcription Features**: For enhancing transcription performance and accuracy, you might want to adjust settings like

region

for compliance or

receivePartialTranscripts

for real-time processing. Check more features [here](https://docs.vapi.ai/changelog/2025/9/8). These updates aim to provide better control and customization over your audio and voice settings. Let me know if there's anything specific you'd like to adjust or need further information on.