Build An Audio AI App Transcripts
Chapter: Welcome to the Course
Lecture: AI +'s and -'s
Login or
purchase this course
to watch this video and the rest of the course contents.
0:00
We're all familiar with large language models and advanced machine learning that have really come on the scene
0:07
in a huge way in the last couple of years. Now, some of these features can be kind of frivolous. Like here's a funny joke.
0:15
Imagine this, this cool guy over there, he thinks he's beat the system. He's like, you know what? I gotta send these business emails.
0:22
They gotta be real formal, but I don't wanna write a lot. I don't wanna write those formal emails. So here's what I'm gonna do.
0:28
I'm gonna write two bullet points, and then I'm gonna hit a button and say, Hey AI, flesh this out, make it sound real professional
0:35
and send it off, okay? Off it goes, lands over on someone's desk. They come back and check their email. They're like, ""Hmm, really? All of this?
0:46
I gotta read all of this stuff? You know what? I don't need this. I got AI, I'm gonna push a button. It's gonna reduce this down to two bullet points
0:54
and tell me exactly what I need instead of all this fluff, right? Maybe we should just learn how to communicate better
0:59
and no one has to go through this, right? So a lot of AI features, they feel like this. However, there are areas where AI machine learning
1:08
really, really do work super well in constructive ways. One of those ways is audio. So imagine we've got some audio file like a podcast, for example,
1:22
and we apply machine learning to it, maybe through Assembly AI's platform. And what do we get back out? Yeah, it's interesting.
1:30
Mark Shannon definitely has vision. He's developed a plan as of years ago, but we finally were able to put him in a position
1:37
where he could do something about it. And we've all been pitching in. A lot of it has to do with just applying
1:42
some of the general ideas that are out there regarding dynamic language and optimization. Things have been applied to other things like LLVM
1:51
or various JavaScript runtimes and so on. So here's a transcript encoded through Assembly AI in the Talk Python from the Talk Python to Me podcast.
2:01
That is word for word perfect of what was said on that show. How incredible is this? It truly is magical. Now, if we take this,
2:11
maybe mix in a little large language models, then we can start to do really interesting things with machine learning and audio.
2:19
And this is the kind of stuff we're gonna explore, these useful AI features in this course.