ghanaopenai/kumawood-speech-transcriptions-v2
Kumawood Speech Transcriptions v2 (Noise-Removed) This is Version 2 of the ghanaopenai/kumawood-speech-transcriptions dataset. In this version, all 194,635 audio segments (~250 hours) have been enhanced using NVIDIA-NemotronLabs-RE-USE, a state-of-the-art multilingual universal speech enhancement model, to remove background noise, room acoustics, hiss, and environmental degradations while strictly preserving speech fidelity and speaker identity. Improvements over v1… See the full description on the dataset page: https://huggingface.co/datasets/ghanaopenai/kumawood-speech-transcriptions-v2.
| Author / organization | ghanaopenai |
|---|---|
| License (as declared) | cc-by-nc-4.0 |
| Downloads | 1.9K (as reported by Hugging Face) |
| Likes | 0 |
| Created | 23 October 2026 |
| Last updated at source | 24 October 2026 |
Popularity, last 4 days
1,895 now (+1,895 since 2026-10-23). Counts as reported by the source, snapshotted daily by Silytics.
More from ghanaopenai
Silytics shows public metadata from Hugging Face and links to the original. Information may be incomplete or out of date. Last refreshed 26 October 2026. Licenses and terms of the original apply to the work itself. Not financial, legal or professional advice.