edge-tts VS crowdcast

i would very much like to use it to turn the text parts of a book into an audio where i could listen to it while reading. i used edge's tts for speech by giving a paragraph to clipboard and to edge-tts in order to listen the text but it causes two problems: 1. you need internet connection and have the book opened 2. can only do paragraph by paragraph, and is prone to errors or sometimes if you use it too much it wont convert the full text afterwards.

This sounds ambitious. I can recommend Edge TTS. Sounds good, albeit with some misspellings. You can see the code here: https://github.com/jing332/tts-server-android Here: https://github.com/rany2/edge-tts Edge TTS is not the only solution, but among the multilingual TTS you will hardly find something better.

I became aware of it through the Home Assistant integration of edge-tts. You can install edge-tts with pip in Termux. For the audio output I also installed 'mpv' in Termux.

Try making a podcast - https://github.com/AdmTal/crowdcast

I recently abandoned a similar project, feel free to steal code if you want - https://github.com/AdmTal/crowdcast

All of the code is here, but I'm going to give some quick links to the prompts below.

As always, the script is on Github - if you're more of a reading type of person 👀

Otherwise, it’s pretty simple, I have a script that pulls the top three comments every week, and I run them through a series of prompts in order to build a full podcast script. Then I use Eleven Labs to generate the voice. I stitch it all together with some music using Python, and then publish it using Buzzsprout. I’m currently using GPT-4 to run the prompts.

Of course, all of the code is open source if you’re into that kind of thing - feel free to open a PR, would love to collaborate and get feedback.