New Greek voice: el_GR-joy-medium #216
Efs-O
started this conversation in
Show and tell
Replies: 2 comments 2 replies
-
|
Sound really better than the default rapunzelina but still needs work compared to english for example. Do you want any feedback? Is this a work in progress? I tried to create some podcasts for personal use just to test it. I can share it if you want. |
Beta Was this translation helpful? Give feedback.
2 replies
-
|
Thank you for you contribution! 🙂 I've added it to the main Piper repo here: https://rhasspy.github.io/piper-samples/#el_GR-joy-medium |
Beta Was this translation helpful? Give feedback.
0 replies
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Uh oh!
There was an error while loading. Please reload this page.
Uh oh!
There was an error while loading. Please reload this page.
-
Hello Everyone! I'd like to share a new Greek TTS voice for Piper — JOY (el_GR-joy-medium).
Voice details:
Language: Greek (el_GR)
Quality: medium (VITS)
Speaker: Chara Kaltsou — native Greek speaker, BA in German Language & Philology (Aristotle University of Thessaloniki), MA (Hellenic Open University)
Sample rate: 22050 Hz
License: CC BY-NC 4.0
Training:
3,216 utterances, all human recordings — no synthetic audio used
Trained with piper-train, fine-tuned from rhasspy/piper-checkpoints (el_GR/rapunzelina/medium)
Dataset covers children's speech, everyday conversation, numbers/dates, school vocabulary, Greek cultural references
Motivation:
Existing open-source Greek voices were trained on synthetic or low-quality data, producing robotic output unsuitable for educational or assistive use. JOY was created to fill that gap — designed primarily for Greek-language educational assistants.
Files:
👉 https://huggingface.co/Efso/joy-greek-tts
📁 https://github.com/Efs-O/Gemma4GR
Audio samples included in the repo.
Attribution required: "JOY Greek voice — Gemma4GR project, CC BY-NC 4.0"
Beta Was this translation helpful? Give feedback.
All reactions