Digital Umuganda
company · RwandaScores
1 product on the map — 1 open-ish.
Openness
3 medium confidence- license
- cc-by-4.0 (card metadata on eight repositories; the Swahili instruction-format copy states none)
- access
- auto (eight repositories behind an automatic Hugging Face gate; the instruction-format copy is ungated)
- dataset_card
- present (current releases describe domains, hours and fields; the Kinyarwanda old version is a placeholder)
Most repositories open after an automatic click-through on Hugging Face, and their metadata carries an attribution-only license. The Swahili instruction-format copy has neither a license nor a card.
- https://huggingface.co/api/datasets/DigitalUmuganda/Afrivoice recorded 2026-09-24
"gated": "auto"; cardData license "cc-by-4.0".
- https://huggingface.co/api/datasets/DigitalUmuganda/Afrivoice_Swahili-Voice_Instruct_Format recorded 2026-09-24
"gated": false; no license in cardData.
- https://huggingface.co/datasets/DigitalUmuganda/Afrivoice_Kinyarwanda_old_version recorded 2026-09-24
Card sections read "[need more information]"; "All datasets are licensed under the Creative Commons license (CC-BY-4)."
- https://huggingface.co/datasets/DigitalUmuganda/Afrivoice_Swahili recorded 2026-09-24
Card: "an image prompt ASR dataset for Swahili... collected on 5 domains"; per-domain hours, fields and loading instructions.
- https://huggingface.co/datasets/google/WaxalNLP recorded 2026-09-24
Dataset Curation table: "Digital Umuganda AfriVoice CC-BY-SA 4.0".
Adoption
2 high confidenceHugging Face downloads summed over all nine Afrivoice repositories, most of them behind an automatic gate. Downloads count file fetches; audio reused inside WAXAL is counted there, not here.
- https://huggingface.co/api/datasets/DigitalUmuganda/Afrivoice recorded 2026-09-24
266 downloads in the trailing 30 days for DigitalUmuganda/Afrivoice
- https://huggingface.co/api/datasets/DigitalUmuganda/Afrivoice_Ethiopia recorded 2026-09-24
134 downloads in the trailing 30 days for DigitalUmuganda/Afrivoice_Ethiopia
- https://huggingface.co/api/datasets/DigitalUmuganda/Afrivoice_Kinyarwanda recorded 2026-09-24
208 downloads in the trailing 30 days for DigitalUmuganda/Afrivoice_Kinyarwanda
- https://huggingface.co/api/datasets/DigitalUmuganda/Afrivoice_Kinyarwanda_old_version recorded 2026-09-24
4 downloads in the trailing 30 days for DigitalUmuganda/Afrivoice_Kinyarwanda_old_version
- https://huggingface.co/api/datasets/DigitalUmuganda/Afrivoice_legacy recorded 2026-09-24
11 downloads in the trailing 30 days for DigitalUmuganda/Afrivoice_legacy
- https://huggingface.co/api/datasets/DigitalUmuganda/Afrivoice_Swahili recorded 2026-09-24
214 downloads in the trailing 30 days for DigitalUmuganda/Afrivoice_Swahili
- https://huggingface.co/api/datasets/DigitalUmuganda/Afrivoice_Swahili_0.0 recorded 2026-09-24
12 downloads in the trailing 30 days for DigitalUmuganda/Afrivoice_Swahili_0.0
- https://huggingface.co/api/datasets/DigitalUmuganda/Afrivoice_Swahili-Voice_Instruct_Format recorded 2026-09-24
107 downloads in the trailing 30 days for DigitalUmuganda/Afrivoice_Swahili-Voice_Instruct_Format
- https://huggingface.co/api/datasets/DigitalUmuganda/Afrivoice_V2 recorded 2026-09-24
292 downloads in the trailing 30 days for DigitalUmuganda/Afrivoice_V2
Capability
4 medium confidenceAfrivoice gives African languages from Kinyarwanda and Swahili to Amharic and Wolof image-described and scripted speech across five releases, more than any other open collection for most of them, and Digital Umuganda's Mbaza recognizer is trained on it while Google's WAXAL corpus draws on it, though it has release cards but no paper. Two releases transcribe only a fraction of their audio, and all five together hold a small fraction of the hundreds of thousands of transcribed hours in the largest English speech corpora.
- https://huggingface.co/datasets/DigitalUmuganda/Afrivoice recorded 2026-11-12
"when available, the transcription of the audio file"; total and transcribed hours: Shona 574.16 and 100.00, Lingala 517.13 and 100.98, Fulani 527.45 and 102.00, Malagasy 516.21 and 102.42, Wolof 530.74 and 102.96, Somali 535.51 and 104.54.
- https://huggingface.co/datasets/DigitalUmuganda/Afrivoice_Ethiopia recorded 2026-11-12
Amharic, Afaan Oromo, Sidama, Wolaytta and Tigrinya, each in unscripted, scripted and expert rows; scripted rows fully transcribed, the others partly; no total row.
- https://huggingface.co/datasets/DigitalUmuganda/Afrivoice_Kinyarwanda recorded 2026-11-12
Total 3162.88 hours, 3155.64 transcribed; models trained on it include DigitalUmuganda/Mbaza-ASR-Afrivoice-660h.
- https://huggingface.co/datasets/DigitalUmuganda/Afrivoice_legacy recorded 2026-11-12
The same six-language table as the first multilingual release, Shona 574.16 total and 100.00 transcribed hours first.
- https://huggingface.co/datasets/DigitalUmuganda/Afrivoice_Swahili recorded 2026-11-12
"This is an image prompt ASR dataset for Swahili."; total 3,217.56 hours, 3,096.87 transcribed.
- https://huggingface.co/datasets/DigitalUmuganda/Afrivoice_Swahili_0.0 recorded 2026-11-12
An earlier Swahili version: total 414.42 hours, all transcribed.
- https://huggingface.co/datasets/DigitalUmuganda/Afrivoice_V2 recorded 2026-11-12
Kirundi, Ndau, Ndebele, Oshiwambo and Gamo table, total 2,421.29 hours, all transcribed.
- https://huggingface.co/datasets/google/WaxalNLP recorded 2026-11-12
source_datasets in the card include DigitalUmuganda/AfriVoice.