Transient Based Assistant - Issue
# support
b
Getting constantly error "Couldn't get assistant" This my webhook response (using Make.com):
Copy code
[
    {
        "body": "{  \"assistant\": {    \"name\": \"Ava\",    \"voice\": {      \"voiceId\": \"248be419-c632-4f23-adf1-5324ed7dbf1d\",      \"provider\": \"cartesia\"    },    \"model\": {      \"model\": \"gpt-3.5-turbo\",      \"messages\": [        {          \"role\": \"system\",          \"content\": \"Ava is a sophisticated AI training assistant, crafted by experts in customer support and AI development. Designed with the persona of a seasoned customer support agent in her early 30s, Ava combines deep technical knowledge with a strong sense of emotional intelligence. Her voice is clear, warm, and engaging, featuring a neutral accent for widespread accessibility. Ava's primary role is to serve as a dynamic training platform for customer support agents, simulating a broad array of service scenarios—from basic inquiries to intricate problem-solving challenges.\"        }      ],      \"provider\": \"openai\",      \"temperature\": 0.7    },    \"recordingEnabled\": true,    \"firstMessage\": \"Hello, this is Ava. How may I assist you today?\",    \"voicemailMessage\": \"You've reached our voicemail. Please leave a message after the beep, and we'll get back to you as soon as possible.\",    \"endCallMessage\": \"Thank you for contacting us. Have a great day!\",    \"transcriber\": {      \"model\": \"general\",      \"language\": \"en\",      \"provider\": \"deepgram\"    },    \"clientMessages\": [\"conversation-update\", \"function-call\"],    \"serverMessages\": [\"hang\", \"model-output\"],    \"endCallPhrases\": [\"goodbye\"]  }}",
        "status": 200,
        "headers": [
            {
                "key": "Content-Type",
                "value": "application/json"
            }
        ]
    }
]
v
@bauman It looks like you've set up your webhook response correctly with a JSON body that defines the assistant's attributes. However, if you're constantly receiving the error "Couldn't get assistant," it could be due to a few different reasons. Let's check a few things: 1. **Assistant ID**: Ensure that the
voiceId
you've provided is correct and that the assistant with this ID exists in the system. 2. **Configuration**: Verify that the assistant configuration is set up correctly in your VAPI system and that there are no typos or errors in the JSON structure. 3. **System Availability**: There might be temporary issues with the VAPI system. Check the system status to ensure everything is operational. 4. **Webhook Response**: Make sure that the response from your webhook is being sent correctly and that the VAPI system can parse it without issues. If you've checked all these points and the issue persists, it might be a good idea to reach out to support for a more detailed investigation. They can provide specific guidance based on your setup and the logs they have access to. Would you like me to guide you on how to contact support, or is there anything else I can assist you with?
You can continue this conversation by mentioning me in the message.
c
Yea that is wrong. not sure where you got that. make a call from the back end and use that json @User
b
I tryed POST
https://api.vapi.ai/call/web
with the same Body object It works without any issues still not clear what is wrong
s
Send me the webhook link in DM.
let me hit the endpoint.
7 Views