When I create new assistant, choose vapi as a provider, attach my personal workflow, it just doesn't work. It just says hello, and doesn't even give any response when I start talking. I tried to recreate the workflow, but nothing. I also tried different provider, and it works, it can hear me and talk. Can anyone help me?

Hey there! 👋 I understand how frustrating it can be when your assistant isn't responding properly. While I don't have access to your specific workflow details, I can definitely help troubleshoot why your Vapi assistant might only be saying "hello" without responding further. Based on the information I've found, there are several potential causes for this issue: ### Transcription Provider Issues One of the most common causes is problems with the transcription provider. According to [user reports](https://discord.com/channels/1211482211119796234/1211483291191083018/threads/1343612796717371393), changing the transcription provider often resolves this issue: - Try switching from Deepgram to Whisper, Gladia, or Azure - One user specifically mentioned: "Changed the transcriber to Azure, that worked for me perfectly" ### Configuration Adjustments You might need to adjust specific settings in your assistant configuration: - Try modifying the

start_speaking_plan

and

transcriber

settings as [suggested here](https://discord.com/channels/1211482211119796234/1211483291191083018/threads/1338497725917827196):

json
  "start_speaking_plan": {
    "waitSeconds": 0.8,
    "smartEndpointingEnabled": true
  }

json
  "transcriber": {
    "model": "nova-2",
    "language": "en",
    "provider": "deepgram",
    "endpointing": 300
  }

### Basic Configuration Checks Make sure these essential configurations are correct: - Verify the assistant ID is correctly set on your phone number - Check that your API credentials for Vapi are valid - Ensure your server URL and webhook configurations are correct ### General Troubleshooting Some additional steps you can try: - Check your VAPI logs for any error messages or warnings - Look at the debugging artifacts sent to your server - Test the assistant in a different environment to rule out network issues - Make sure your workflow doesn't have any script errors

Hey! To help track down this issue, could you share: - The call ID - When exactly this happened (the timestamp) - What response you expected to get - What response you actually got instead This would really help us figure out what went wrong!