Hacker Newsnew | past | comments | ask | show | jobs | submitlogin

macOS has a very good built-in text-to-speech engine with real voices that sound good (i.e. not annoying to read).

It's part of the Accessibility > Speech menu in Settings.

I wrote up a little info doc for how to enable keyboard shortcut for it so you can use with any text you can select: https://docs.google.com/document/d/1mApa60zJA8rgEm6T6GF0yIem...

Use cases: - read blog posts and news articles while doing dishes (basically turn anything into an podcast at the press of a button) - read hacker news comments - proofreading text (you can hear the typos much easier than you can see them in a text you have written)



I'm even creating my own audiobooks from ebook text, as a way to consume books not officially available in audiobook!

I haven't tested it much yet but the 'Samantha' voice actually seems listenable and intelligible for long periods.


Do you think there will come a time when audiobooks are generated via sequence model using celebrity voices?

I gave it a go with LSTM's.

Each sentence is intelligible, but it feels so dry without the emotion of a human voice actor.




Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: