Mastodawn

So, it seems #PiperTTS is losing steam as no one seems to know much about it on here. So, what *are* people using these days for clonable voices that actually work with #NVDASR? Or is Piper the be-all-end-all?

Tarren (They/Them)5d ago

Any ideas why this voice won't work in #SonataNeuralVoices? I tarred both the .onnx and .json file, but Sonata claims the voice failed to load for some reason even though it works fine in #PiperTTS.
https://github.com/dividebysandwich/piper-voice-models/tree/main/Data
If you don't know, please boost for reach.

piper-voice-models/Data at main · dividebysandwich/piper-voice-models

Pre-trained and ready to use voice models for Piper TTS - dividebysandwich/piper-voice-models

GitHub

Tarren (They/Them)Jun 14

OK, lo,oks like today is the day we find out if I've made something that can actually be a decent #PiperTTS voice, or if I 've totally screwed up...

Tarren (They/Them)Jun 12

Well, I am over halfway through my #PiperTTS voice-cloning process. Only reason I'm typing this right now and not recording is because our neighboring meat market's delivery truck just *HAD* to show up right in the middle of things.

***Additional Hashtags***
#Blind #Accessibility #A11y #TTS #Piper #VoiceCloning

Tarren (They/Them)Jun 12

YAY!! I got #PiperRecordingStudio and my mic to play nice with each other finally; I'm 286 sentences into training. If I get a voice I'm happy with, I'll be happy to share it as an #NB voice by an NB speaker.

***Additional Hashtags***
#PiperTTS #Piper #TTS

oxytude May 3

Cette semaine sur Oxytude, Podcast sonnette connectée #Blink et actu des nouvelles technologies et de l'#Accessibilité.

#SuperNova #BookWorm #Soundscape #Android #Slack #PiperTts #Pages #macOS #TweeseCake #Kindle #DRM #XChat #APIDV #ImpotsGouvFr

https://www.oxytude.org

Oxytude – Accessibilité, informatique et nouvelles technologies

Reddit Tech VN Bot Nov 2, 2025

"Khó khăn với huấn luyện giọng nói custom Piper TTS, lỗi xung đột sau khi hạ cấp gói. Đang tìm Docker Container hỗ trợ và hướng dẫn! #PiperTTS #GiọngNóiTuỳChỉnh #Docker #CustomVoice #CôngNghệ #AI"

https://www.reddit.com/r/selfhosted/comments/1omiv5a/piper_tts_custom_voice/

Show thread

chibi-[N]ah🇫🇷

Jun 21, 2025

Piper-TTS :

https://github.com/rhasspy/piper

Voix style GlaDOS :

https://github.com/TazzerMAN/piper-voice-glados-fr

#piperTTS #piper #tts #GlaDOS

GitHub - rhasspy/piper: A fast, local neural text to speech system

A fast, local neural text to speech system. Contribute to rhasspy/piper development by creating an account on GitHub.

GitHub

Show thread

chibi-[N]ah🇫🇷

Jun 21, 2025

Test un peu plus sérieux.

Commande utilisée

echo "Enfin, que dis-je, enfin, finalement, une synthèse vocale avec une voix française qui prononce les mots de manière intelligible ! Ça change tellement des voix sans prosodie !" | ./piper --model voices/fr_FR-upmc-medium.onnx --output-wav synt.wav

#piperTTS #TTS

chibi-[N]ah🇫🇷

Jun 21, 2025

Quitte à utiliser une IA, autant utiliser la voix de GlaDOS.

Enfin une synthèse vocale (TTS) avec une voix en français qui fonctionne et est intelligible.

Autrement dit, juste un prétexte pour tester piper-tts

Commande utilisée

echo "Quitte à utiliser une IA ; autant utiliser la voix de Gla DOSSE." | ./piper --model voices/fr_FR-glados-medium.onnx --output-raw | aplay -r 22500 -f S16_LE -t raw -D pipewire -

#piperTTS #TTS