ghanaopenai/ghana-voice-training-data

Ghana Voice training data The exact clips each Ghana Voice language model was fine-tuned on, one subset per language (ghanaopenai/ghana-voice_<subset>). All audio is in one voice, clear-high-slow from africa-female-vc: either taken from ghana-speech-female-22khz_bible / ghana-speech-female-22khz, or, where a language had too little, converted from ghana-speech recordings into the same voice. Each clip passed the training filters: no digits in the text, 3-15 s, silence trimmed… See the full description on the dataset page: https://huggingface.co/datasets/ghanaopenai/ghana-voice-training-data.

<iframe src="https://silytics.si/embed/item/feba15ab-3886-4c3e-95d5-354f3ea6a786" width="420" height="190" style="border:0" loading="lazy" title="Silytics entry"></iframe>

Silytics shows public metadata from Hugging Face and links to the original. Information may be incomplete or out of date. Last refreshed 24 October 2026. Licenses and terms of the original apply to the work itself. Not financial, legal or professional advice.