Lucas van Staden
09/10/2025, 5:56 AMVapi Support Bot
09/10/2025, 5:57 AMGEORGE || AI AND WEB INNOVATOR
09/10/2025, 11:27 AMChiranjeet Mishra
09/11/2025, 2:59 PMsilence-timed-out reason is documented in the call logs.
2. **Assistant-Related Issues**: Common causes include the assistant recognizing a specific phrase that ends the call, or encountering an error such as assistant-request-returned-no-assistant or assistant-said-end-call-phrase.
3. **Call Duration**: If the call duration has reached the maximum limit, it might terminate as per the exceeded-max-duration reason.
For a detailed list of call end reasons, you can refer to the [Call Ended Reasons documentation](https://docs.vapi.ai/calls/call-ended-reason).
I recommend reviewing your call logs to identify the specific reason. Adjustments can be made based on that insight, such as modifying assistant prompts or settings.Lucas van Staden
09/11/2025, 9:24 PMLucas van Staden
09/11/2025, 9:24 PMChiranjeet Mishra
09/15/2025, 8:41 PMivrig
09/26/2025, 8:09 AMLucas van Staden
09/28/2025, 12:36 AMLucas van Staden
09/28/2025, 12:38 AM"fallbackPlan": {
"transcribers": [
{
"provider": "assembly-ai",
"language": "en",
"confidenceThreshold": 0.4,
"formatTurns": true,
"endOfTurnConfidenceThreshold": 0.7,
"minEndOfTurnSilenceWhenConfident": 160,
"wordFinalizationMaxWaitTime": 160,
"maxTurnSilence": 400,
"realtimeUrl": "string",
"wordBoost": [
"string"
],
"endUtteranceSilenceThreshold": 1.1,
"disablePartialTranscripts": true
}
]
}
ref: https://docs.vapi.ai/api-reference/assistants/createLucas van Staden
09/28/2025, 1:06 AM{
"transcriber": {
"model": "gpt-4o-transcribe",
"language": "en",
"provider": "openai",
"fallbackPlan": {
"transcribers": [
{
"provider": "11labs",
"language": "en",
"model": "scribe_v1"
}
]
}
}
}ivrig
09/28/2025, 11:42 PMLucas van Staden
10/04/2025, 12:46 AM"transcriber": {
"model": "gpt-4o-transcribe",
"language": "en",
"provider": "openai",
"fallbackPlan": {
"transcribers": [
{
"model": "scribe_v1",
"language": "en",
"provider": "11labs"
}
]
}
},
test suite fail
The evaluation outcome cannot be determined because the call transcript is missing. Without the transcript, it's impossible to assess whether the AI successfully identified the patient by phone number and called them by their first name, created a minor account, made a new booking for the minor, and maintained polite and professional responses. Therefore, the test is marked as failed due to a lack of information.
all transcripts seem to be failing.Lucas van Staden
10/04/2025, 12:49 AMLucas van Staden
10/04/2025, 1:08 AMChiranjeet Mishra
10/07/2025, 2:19 AM