Hey All. I have a squad running for a client and when I first set it up I was able to stop it from saying that it was transferring from one assistant to the next as it is in essence the same person just with different skill sets & knowledge. Previously I was able to get it to not say that it was transferring to the xyz specialist. But now we have added a human transfer to an external number so it is now explaining all transfers. It can see why the AI is getting confused as the transfer is a squad is called a transfer call the same as transfer to external number. Does anyone have any prompting suggestion where you can force the AI to stop doing this?

I'm looking for some advice on outbound call structure. I have set the "firstMessageMode": "assistant-waits-for-user" so that the AI doesn't talk over someone answering the call. It seems that I do however need a first message to start the AI talking. The issue with that is the first message is not responding directly to what the user just said. To clarify... my issues or scenario is based on if I'm talking to the correct person or not. This can be defined by the user answer the call with "Hi, this is XYZ." The AI can then cross check the dynamic field that it is the same and therefore continue. If the user doesn't answer with their name. It would makes sense to ask at some point "Am I speaking with XYZ?" The other part is that it would seem that after the first message is spoken the AI waits... it can't continue without a response. So it can't ask follow up questions like am I speaking with... etc. Am I over thinking this and I should just continue assuming that you are talking to the correct person (These are Warm leads)... OR doesn't someone have some advice or examples of what they are using.

I have a question about training Deepgram or alternatives for accuracy when it comes to addresses. I want to load up all the suburbs for the area our client is working in so that it knows what to pull from. I know Deepgram you can add 100 keywords... But ideally, I want more than that. I want 500 or so. And I'm cautious that adding even 100 will cause latency (Plus how do I know which are the top 100 to add). We are already verifying the address via Google but if the transcription falls over it can be a 2 or 3 attempt to get some more local pronunciations across the line. Any thoughts would be appreciated.

but in the mean time, this may be helpful for word pronunciations: // CHARACTER PRONUNCIATION MAPPING DEFINE CHARACTER_PRONUNCIATION { "A": "ay", "B": "bee", "C": "see", "D": "dee", "E": "ee", "F": "eff", "G": "gee", "H": "aitch", "I": "eye", "J": "jay", "K": "kay", "L": "ell", "M": "em", "N": "en", "O": "oh", "P": "pee", "Q": "queue", "R": "arr", "S": "ess", "T": "tee", "U": "you", "V": "vee", "W": "double you", "X": "eks", "Y": "why", "Z": "zed", "0": "zero", "1": "one", "2": "two", "3": "three", "4": "four", "5": "five", "6": "six", "7": "seven", "8": "eight", "9": "nine", "/": "forward slash", "-": "dash", ".": "dot" }

**Question for the group** Who is having issues creating and managing client phone numbers.... My hypothesis is when a new client is onboarded you have a process to either: - Buy a number using your own Twilio account (or other) - Setting them up on a sub account, within your account then getting identity documents and submitting - Asking them to set up a twilio account and do the steps from the previous then getting access or account details... pain. - Porting their exciting number to your account so you can manage it. Each of these elements have pros, cons and pain. [Let me know if I've missed a point] Other issues will be diverting from one number to the next ans when... bit this can be seperate. Let me know if this is your pain and which you are favouring... and maybe I'll fix it 😁