Free and Open Source Software

18535 readers

33 users here now

If it's free and open source and it's also software, it can be discussed here. Subcommunity of Technology.

This community's icon was made by Aaron Schneider, under the CC-BY-NC-SA 4.0 license.

founded 2 years ago

MODERATORS

Gaywallet@beehaw.org

alyaza@beehaw.org

Open Source Text-to-Speech and Speech-to-Text on Android? (lemmy.dbzer0.com)

submitted 1 day ago* (last edited 1 day ago) by andrew0@lemmy.dbzer0.com to c/foss@beehaw.org

13 comments fedilink hide all child comments

Hello everyone! I am interested in replacing the Google Speech Recognition and Synthesis app on Android. For Speech-to-Text (STT), I've tried Whisper and FUTO, and settled on the latter because it seemed to be more versatile. Also, FUTO seems to have some decent recognition, but not yet capable of handling all the languages that I want. Regardless, so far happy with STT. The only annoyance I have is that it does not appear as an option in the settings for Speech recognition :(

However, I can't seem to find any replacements that have good Text-to-Speech (TTS) quality. I tried espeak-ng and RHVoice, but both have robotic outputs.

Given the recent advancements in AI, I was expecting that there would be ways to incorporate open source TTS models like Kokoro to generate speech on the go. Nevertheless, I could not really find any such apps so far.

Has anyone managed to completely replace the Google app with (an)other privacy-focused FOSS app(s)?

top 13 comments

sorted by: hot top controversial new old

[–] infeeeee@lemm.ee 8 points 1 day ago* (last edited 1 day ago) (2 children)

Sherpa onnx is very good: https://k2-fsa.github.io/sherpa/onnx/tts/index.html

Apks are here: https://k2-fsa.github.io/sherpa/onnx/android/apk.html

[–] sic_semper_tyrannis@lemmy.today 5 points 1 day ago

Sherpa is by far the best. I personally find GB southern English female medium very natural sounding

[–] andrew0@lemmy.dbzer0.com 2 points 1 day ago* (last edited 1 day ago)

Thanks! I was actually looking at this, but I gave up because I couldn't really figure out how to get a multilingual model running through Obtainium. I'll try again :D

[–] mustbe3to20signs@feddit.org 5 points 1 day ago (2 children)

I (rather seldom) use FUTO Voice for STT and SherpaTSS for TSS.

[–] andrew0@lemmy.dbzer0.com 1 points 1 day ago* (last edited 1 day ago) (1 children)

Thanks for the SherpaTTS suggestion. I really like the GLaDOS voice <3

I am not sure which phone you use, but are you able to set FUTO Voice as the default "Voice input" in the Android settings? I played around with a few apps, which show up. However, FUTO is not an option here :(

[–] mustbe3to20signs@feddit.org 2 points 5 hours ago

Honestly, idk. Anysoft Keyboard and Firefox integrated seamlessly, so I never really checked on system level.
Settings says there is no app set up, but if I tap it, it is listed as the only choice?

[–] Pirata@lemm.ee 3 points 1 day ago* (last edited 1 day ago)

My setup exactly. I can highly recommend this approach.

[–] Ulrich@feddit.org 4 points 1 day ago (2 children)

Transcribro

[–] Showroom7561@lemmy.ca 4 points 1 day ago

Thanks for that. It works quite well.

Even though I'm using FUTO (paid for it), I don't mind trying new things. I just wish these produced text in real-time as I sometimes dictate long-form content, and it's unnerving to wait a while for it to process (not knowing if it actually heard everything).

[–] andrew0@lemmy.dbzer0.com 2 points 1 day ago (1 children)

Thanks for the suggestion! I gave this a try, but it seems that it won't register any voice 🤔 However, it seems like it shows up in my settings, so it's a good sign. I'll try to get it to work :D

[–] Ulrich@feddit.org 3 points 1 day ago

Weird! Worked fine out of the box for me!

[–] said@lemmy.sdf.org 4 points 1 day ago (1 children)

I'm using one of the models here . It's working for voice instructions for navigation in Organic Maps. It's also showing in the text to speech menu in the android settings.

The app it's based on (SherpaTTS) already exists on fdroid repos, but the APKs on that link has the model embedded already.

[–] Cris16228@lemmy.today 3 points 1 day ago

:O that's what I was looking for and I forgot about this website! If it works, I'm going to hug you :c