Logo Lanfrica
  • Home
  • Atlas
  • Insights
  • Docs
  • Sign in

© 2026 Lanfrica. All rights reserved. All copyrights of the resources shown on the Lanfrica website belong to the original copyright holders, unless explicitly stated otherwise.

vysakh25/kinyarwanda-radio-tts

Domain:

natural language processing

Record type:

dataset
Creator:
vys
Host:
Radio Rwanda broadcast data with SNAC audio tokens for Orpheus TTS training. Source: Radio Rwanda (RBA) via vysakh25/kinyarwanda-radio-tts-raw Clips: 19,145 (6 skipped for length/empty tokens) Dominant speaker: Ismael Mwanafunzi (~80%, 27.6h) Quality: STOI 0.987, PESQ 3.11, SI-SDR 22.66 SNAC codec: hubertsiuzdak/snac_24khz (7-token interleave)

Visit

huggingface.co

Tasks

speech processingtext to speech

Languages

Kinyarwanda

Similar

vysakh25/piper-tts-kinyarwanda-mediumKinyarwanda TTS ModelKinyarwanda TTS Dataset

vysakh25/piper-tts-kinyarwanda-medium

Kinyarwanda TTS Model

Text-to-Speech model for Kinyarwanda

Kinyarwanda TTS Dataset

Dataset made of more than 17 hours of Kinyarwanda studio recordings