Hi HN,<p>I wanted to get back on track with the current machine learning literature. When I found the free book "Understanding Deep Learning," I had the idea to listen to the book instead of reading it on a screen.<p>So I've built a tool to turn PDFs into MP3s so I can listen to it. I'm now almost done with the book; I listen to the audio chapters every day on my commute to work.<p>For regular text, you can simply use any text-to-speech API, but for technical text containing formulas, it's more contrived. I'm using ChatGPT to simplify the text for a better audio experience. Listening to formulas while driving in the car just doesn't work for me. When there is a formula in the text, I try to describe it like "... a linear combination of the terms ...". It's work in progress.<p>I've just launched the MVP, so let me know what you think. Do you like the simplification of the text? <a href="https://www.pdftomp3.com/pdfs" rel="nofollow">https://www.pdftomp3.com/pdfs</a><p>P.S. the pricing is due to the API calls.
I'm planning to include a sample page featuring a paper available for download in MP3 format. I was considering "Attention Is All You Need". Are there any other papers you'd like to have in audio form?