I have a call where AI asked for invoice#. While providing invoice#, AI interrupted the user. Is there any settings can i change to prevent this from happening? Call ID: 019b6c61-8e12-799c-beb9-d0f3f631b201+ You can check the log from 20 seconds to 40 seconds.

To prevent the AI from interrupting users while they provide information like an invoice number, increase the assistant's wait time before responding. Add or update the

waitSeconds

parameter in your

startSpeakingPlan

to make the assistant more patient:

json
"startSpeakingPlan": {
  "waitSeconds": 1.0
}

This change will give users more time to finish speaking before the AI responds. For more details, see the documentation I have found: [Voice Pipeline Configuration](https://docs.vapi.ai/customization/voice-pipeline-configuration). Source: - [Voice Pipeline Configuration](https://docs.vapi.ai/customization/voice-pipeline-configuration)

I recommend LiveKit endpointing, though nothing is perfect

Livekit is a good choice for smart endpointing, and if that doesn't work for you use Vapi as endpointing. It can be found in the Start Speaking Plan section