FOSS Text To Speech engines?
Hi guys! I'm trying to see if there's alternatives in TTS for the (admittedly) great one Google provides. I'm using Graphene, and so that's not an option I'd like to go around.
Alternatively on the FOSS field, I believe have RHVoice, which I have installed, and eSpeak, which was last updated in 2022 and is not compatible with my Pixel 7 64bit-only phone. So...RHVoice has the problem of having their voices a bit robotic, but more probrematic is that they're VERY muted. On max volume, they're not all that loud, and I ride a motorbike. They're pretty useless, most of the times I can't hear the directions on my headset. Do you guys know any alternative TTS engine I could be using? I've heard the demo sample from the aiyu-ayaan tts-engine and it sounded great, but no idea how can I compile/install it as an Android TTS engine available for other apps to use, just like RHVoice.
Alternatively, are there any other apps/engines you'd recommend?
Thank you!
https://github.com/k2-fsa/sherpa-onnx has some apks you can download. The quality of the voice I've used is fantastic: I prefer it to Google's to be honest. The program itself has been buggy for me, though... Needing to launch the TTS launcher before I can use any application that would use it
Wow, never heard of this one, and it sounds so much better! I'm going to test it around, thanks so much!
Unfortunately as far as I've been able to find, those seem to be two of the main open source options (besides some AI TTS stuff, but I can't seem to find Android versions of those yet). Nevertheless, I did a little digging and found you're not alone in thinking RHVoice is on the quiet side, which led me to someone pointing out that you might try editing the config files and adjusting the volume values there, supposing you haven't already.
The person that submitted the issue never came back to mention whether or not that did the trick for them, so best of luck!
Thanks for the digging! I actually saw that comment and edited those files back in the day, to no visible (audible?) difference. So at least for me, running GrapheneOS, it didn't work.
I use https://gitlab.futo.org/alex/voiceinput
Pretty good for me so far
Thats STT not TTS
My bad, I misread it
We all do :D