I used the appointment scheduling template to test the capabilities and it keeps telling me "I have an appointment at double one one thirty a m" even in the prompt I added a line to say "appointment time 11 am should be read as eleven thirty a m or eleven thirty morning" How do I fix this?

You cannot prompt the audio model to read words - it has to be typed by the LLM for the audio model to speak out. Try adding in your prompt: "Any numbers must be spelled out in words so that it can be read by an audio model." You can even add an example: "for example, 11:30 would be: eleven thirty"

Hey, Can you send me the latest call_id? Along with that, can you provide me the duration in the recording where you had this issue?

I will check this and get back to you, but im not sure how well it would work, like so far the default template you have has also a line for spelling things out, but no matter what nothing is spelled out.

No worries Specialist just send the call ID when you can. Along the lines of Slavi's comment, make sure you're explaining to the LLM how to write text as opposed to "saying" things. It helps to put examples of the expected output such as Instruction(how you want it to output words spelled out -- **Example**(an example or two of words before and after being spelled out(

Yeah I had a similar problem with like pricing which I fixed by writing it all out. I haven't tried any more advanced integration yet, was just using the built in editor. So far I am impressed but there's been times where the audio repeats (I know different issue). Appreciate the help I received, with this main issue I described here, thanks

Awesome man I'm glad I could help you out and I'm happy to hear you're liking the platform. If you create a new ticket with a call ID i'd be happy to dive into it!