ef2a99f3-2aca-4462-a569-a63fa380e098 @shubham.bajaj, please check and resolve this issue, our calls have stopped, 20s call and 0.20 cents charged! Secondly the delay has now increaassedd!! The endpointing issue or whatever it is please. I am waiting.

Same here, what a mess !! It's maybe related to knowledge base Turn latency: 9412ms (transcriber: 298ms, endpointing: 101ms, kb: 7898ms, model: 8249ms, voice: 743ms)

Hey, we will be having an Office Hour in 20 minutes. Can you hop in and get your query resolved during it?

@Shahzaib.aiii The call cost is $0.20 because it's made up of different provider components. Here's the breakdown: - LLM (AI Model Processing) – $0.1388 - This is the biggest chunk of the cost. The call involved 21,864 prompt tokens and 87 completion tokens, processed by OpenAI’s GPT-4o. - Speech-to-Text (STT) – $0.0169 - This covers the transcription of the call using Azure for about 0.5 minutes of speech. - Text-to-Speech (TTS) – $0.0146 - The system converted 292 characters into voice using 11Labs’ Eleven Turbo V2. - Voice API (VAPI) – $0.0169 - This cost comes from handling 0.3372 minutes of audio streaming. - Analysis (Summarization & Data Processing) – $0.0138 - This includes summarizing, structuring data, and evaluating the call's success using Claude 3.5 Sonnet from Anthropic. All of this adds up to $0.201, which rounds off to $0.20. Essentially, most of the cost comes from AI processing, followed by voice-related services.

Shubham we will have to hop on a call to understand thiss!