Showing posts with label artificial intelligence. Show all posts
Showing posts with label artificial intelligence. Show all posts

Wednesday, October 2, 2024

Mind Blowing AI Podcast Creation

This podcast was created for free, entirely by an AI based on approximately 1,200 words of text: http://joemoreno.com/patriotism.mp3

Everyone I’ve shared this with has been absolutely amazed.

Those two voices, in this podcast, are completely AI generated. The source of the text the AI used was this blog post from earlier today.

Try It Yourself

Here's how I did it (you'll need a Google/Gmail account).

1. I visited Google's NotebookLM AI at: https://notebooklm.google.com

2. I clicked the Create button.

3. I then uploaded the text of my document.

4. Under the "Audio Overview" I clicked on the Generate button to create the "Deep dive conversation" with two hosts. It took a few minutes to complete and then I downloaded the WAV sound file. I added one additional, unnecessary, step of converting it to an MP3 using the Apple Music app on my Mac (formerly known as the iTunes app).

Ta-da. That's all there is to it.
Since this is generative AI, every time you use the same content to generate a new podcast, you'll get a different result.

Use Cases

I'm thinking of other uses for this such as turning a long article into a podcast to listen to during a commute or perhaps some work related reading that you can listen to as a podcast while commuting. 

Wednesday, August 28, 2024

Hallucinations: How Many ‘R’s Are in the Word Strawberry?

Ask your favorite AI chatbot, "How many ‘R’s are in the word strawberry?"

Most will respond with, "The word 'strawberry' contains two ‘R’s." Obviously, that's wrong – the correct answer is three.

This is the difference between knowing and understanding. 

AI models tokenize words. Tokenization is the process of breaking down a stream of text, such as sentences, into individual words and then assigning values to each word in multiple dimensions. An AI model doesn't break down a word into letters, so current models don't use introspection to know what letters make up a word. While an AI model could break down words into letters, the juice is not worth the squeeze when it comes to memory and storage requirements. 

In the world of AI, this seemingly confidence, yet random guess, is called a hallucination.