search

rhasspy / piper

A fast, local neural text to speech system
https://rhasspy.github.io/piper-samples/
MIT License
4.38k stars 297 forks source link

Add Direct Links To More Voices #435

Closed tvraman closed 1 month ago

tvraman commented 2 months ago

Could you please add more direct download links for voices in the readme.md file; at present you have one as an example. But the various pages where other models live are a maze of links and is hard to use -- especially if one cannot see

luiscarlos2000 commented 2 months ago

I was asking the same question since it-s beginning. I also need additional repositories a longside the official one.

tvraman commented 2 months ago

the problem is somewhat eased by the data in voices.json and the piper executable's option of --update-voices; could be better documented to say that that works only if the datasets dir is writeable by the user running piper.

Also the Hugging Face site though well-structured to reflect the underlying datastore is a mess to navigate -- wish someone had a programmatic API client for it.

--

synesthesiam commented 2 months ago

I could generate a markdown file with links to all the voices. Something like:

* English (U.S.)
    * Lessac
        * Low (link)
        * Medium (link)
        * High (link)

Would this be any better?

tvraman commented 2 months ago

that would be awesome!

--

synesthesiam commented 2 months ago

It's live now! https://github.com/rhasspy/piper/blob/master/VOICES.md

tvraman commented 2 months ago

Nice, Thanks for all your work on Piper -- it's a very nice piece of work.

--