renamed split_long_sentence to get_sentences, added punctuation for each language into language_mapping, unused code removed

[ROBERT-MCDOWELL]

Please test it with a long chapter and long sentences in 3 different language signs (english,arabic,chinese for ex)
as I cannot do anything with my 16 years old laptop core duo with a 512MB nvidia GPU :-|

[DrewThomasson]

I'll be Testing this later with my

alice_in_wonderland.txt

Along with other translations I make of it for the other languages with chatgpt lol

[ROBERT-MCDOWELL]

I'm working to add all languages in language_mapping... tedious work...

[DrewThomasson]

:o

Anything I could do to help( Whenever I have some time to spare every now and then?)

• Sadly I might not have much time to put into the repo, as finals are coming up lol :(

Oh yeah finding voices for other languages, lol you did ask about that before ☝️ 😅

• I can ask all my multilingual classmates for voice actor suggestions :0

[ROBERT-MCDOWELL]

It's ok I almost finished. 20 by 20 after a chagpt limitations with the right request without hallucinations...
maybe the best would be to test various new languages? does it need to update the code?

[ROBERT-MCDOWELL]

for the voices I have already some.. we will share the job but not a priority for now.. the v2 merge is...

[DrewThomasson]

Your code works in a space too btw! :D

https://huggingface.co/spaces/drewThomasson/ebook2audiobook_v2.0

[ROBERT-MCDOWELL]

Hurrayy! :D... the get_sentences + enable_text_splitting=True is not to bad, but sometimes word repetition and not enough space between some end of sentence and start of the following....

[DrewThomasson]

hm, Well I think the word repetition was already a thing.

That's just a thing that happens with xtts for some reason.

I don't think it happens for fairseq at least lol

[ROBERT-MCDOWELL]

this will be the last thing to solve before the merge... GN!

[DrewThomasson]

kk GN!