Is it just me, but when I am on my iphone and talk to the orb i can tell it's listening to me, but I can't hear anything on my end when it responds. But from the UI it looks like it is responding.... Works perfectly on desktop. Test here: mysunnyai.com

I have looked everywhere (I think?) - I can not seem to find where we can sample/test the voice ai before having to call the number or put it in an orb? My client wants to go through and hear a bunch of voices, is there not a button for this?

If you’re building with Assistable, this is your once-a-year chance to lock in massive discounts on plans, snapshots, and usage. We rolled out three limited-time deals you can stack or choose from depending on what you need: 1️⃣ Annual Plan Deal (Biggest Savings): Get 15 months for the price of 10 on any annual plan + free voice minutes.Perfect if you want long-term savings and more usage runway. https://go.assistable.ai/black-friday-annual-plan-offer 2️⃣ Snapshot Deals: All niche snapshots are 40% off, and you can also grab the full 16-snapshot bundle at a massive discount. Great if you want more done-for-you systems to resell. Also, if you purchase the bundle we will include a fully white-labeled document portal for you. You know our help articles at help.assistable.ai? We will duplicate that for you but with your own branding and domain. 3️⃣ Usage Wallet Deal: Spend $2,000 and get $3,000 in usage credit.Basically $1,000 of free AI Voice minutes. https://go.assistable.ai/black-friday-offer All deals are live now and end soon.If you want to tighten up your offer, scale with AI Voice, or add more niche automations to your agency—this is the best time of the year to do it.

A client only wants to pay for usage, is there anywhere to set up rebilling specifically for the AI tokens for SMS? Standard Assistbale cost is $0.02/token - can I make this like $0.4?

Does anyone know how to ensure this doesn't happen? I have $50 in credits in my Open Ai account. These limits thing are a little odd to understand, they seem different from credits. But how can we avoid these limits... apparently too many tokens were used per minute? Assistable did not respond to a customer because of it. Any insight on this?