Have you ever thought of some crazy idea but didn’t start taking actions ? but we did, we did take action. Me and my co-founder, we thought about an idea we had this vision aligned together in our minds. We researched and again and again about it. And it was not there, literally no one made it how we were thinking. we saw this opportunity and started drilling into the scope and architecture. I will tell you the idea later but right now our main focus is on the components it required to build, we needed a 3d model, we needed AI API tokens, we needed voice AI tokens. I hope, you must have some idea by now, what I am talking about.
Well, being college students we didn’t have capital/resources to build even the MVP for our product to atleast showcase to this world and expect some funding.
we tried though, building our MVP we planned to make it layer by layer. first starting with 3d model, adding voice to it, adding movements would be the later part, as you know resources, haha. So, in our most capable laptops we were trying hard to find the required 3d model and then to render it. Somehow it again gave us the problem of not finding a good 3d model, there was 1 website which provides it for free, but it got closed a few months back, So, we tried the existing ones, but it had no relevance because during testing they had their trial over, even before we start building our MVP. lol, So, we decided to start backwards, we will first work on our AI agent part, then integrate voice to it, then integrate 3d model to it. easy, right!? Not so.
Get an AI API key done, called it, integrated text to speech API (didn’t use voice ai agents they cost money 🤑) So, we built our own, which we can modify and improve as we want, Done. Our basic voice ai agent was ready, Well honestly not voice ai agent but voice ai chatbot. with a button to listen to our voice and respond accordingly. It had some glitches but works totally fine. using principle called one thing at a time. just kidding, we can’t speak to it, during it is speaking. Not, going much in detail, This was the scenario.
By now, you must be sure what we were building, go and comment the guesses,
So, the idea was to build a 3d virtual ai agent, which could be accessed anywhere on any screen just like a cartoon in animated movies, roaming around on your screen and interacting with you in real time. it’s like a real creature but inside a screen. it could perform tasks which will be visible to the user. let say, user say play “senorita” on YouTube, our ai would open a browser by picking it from an app menu and typing on search bar and clicking, all this in real time just talking to it.
crazy!? isn’t it. Well, sometimes things which are good in assumptions are not feasible in real life. I mean not now, atleast. In future, I am definitely sure people will be working on it.
So, that’s the end. I hope you like it. and see you in the next one.