Gabriel C
05/31/2024, 4:13 PMVapi Support
05/31/2024, 4:13 PMVapi Support
05/31/2024, 4:13 PMsilenceTimeoutSeconds
, responseDelaySeconds
, and llmRequestDelaySeconds
are optimized for your use case. Adjust these settings if necessary to ensure that the assistant is correctly processing speech and not cutting off important parts of the conversation [(1)]() .
3. **Transcriber and Voice Settings**: Check the configurations related to your transcriber and voice settings. Make sure that the language, model, and other specific settings align with your requirements.
If these steps do not resolve the issue, it might be beneficial to look into the specific call ID you provided and check server logs for any errors or unusual activity during the time of the calls. If you continue to experience issues, consider reaching out to Vapi support with detailed information about the problem and any relevant call IDs for further investigation.
For more detailed troubleshooting, you might want to delve into the API references for updating or configuring assistants, where you can adjust various parameters that might influence recording and transcription quality.
If you need to adjust these settings programmatically, here's how you might update your assistant settings via API:Vapi Support
05/31/2024, 4:13 PMjson
{
"silenceTimeoutSeconds": 30,
"responseDelaySeconds": 0.4,
"llmRequestDelaySeconds": 0.1
}
Make sure to replace placeholder values with the actual settings that suit your scenario.
If the problem persists, it would be advisable to examine the logs for the specific call ID and check for any error messages or warnings that could give more insight into what might be causing the jumbled recordings and transcriptions.Sahil
05/31/2024, 6:22 PMGabriel C
06/03/2024, 2:39 PMSahil
06/03/2024, 2:40 PM