This collection contains benchmark datasets created for speech recognition for African languages by the UCT NLP research group.
AI & ML interests
We develop language resources and language models for low-resource South African languages
Recent Activity
View all activity
Organization Card
UCT NLP
We are the Natural Language Processing research group at the University of Cape Town.
We develop language resources and models for low-resource South African and African languages. Our research includes low-resource language modelling, South African language data, linguistic and knowledge structure, data-efficient modelling, interpretability, and faithful text generation.
We release datasets, models and code to support open and reproducible research.
The group is led by Dr Jan Buys and Dr Francois Meyer.
OmniASR CTC models (300M, 1B, 3B) fine-tuned for isiZulu speech recognition, with training-data ablations.
-
uctnlp/omniASR-CTC-300m-v2-Zulu-Baseline
Automatic Speech Recognition • 0.3B • Updated • 33 -
uctnlp/omniASR-CTC-300M-v2-Zulu-Lwazi
Automatic Speech Recognition • 0.3B • Updated • 15 -
uctnlp/omniASR-CTC-300M-v2-Zulu-No-Lwazi
Automatic Speech Recognition • 0.3B • Updated • 18 -
uctnlp/omniASR-CTC-300M-v2-Zulu-All-v1
Automatic Speech Recognition • 0.3B • Updated • 12
This collection contains benchmark datasets created for speech recognition for African languages by the UCT NLP research group.
OmniASR CTC models (300M, 1B, 3B) fine-tuned for isiZulu speech recognition, with training-data ablations.
-
uctnlp/omniASR-CTC-300m-v2-Zulu-Baseline
Automatic Speech Recognition • 0.3B • Updated • 33 -
uctnlp/omniASR-CTC-300M-v2-Zulu-Lwazi
Automatic Speech Recognition • 0.3B • Updated • 15 -
uctnlp/omniASR-CTC-300M-v2-Zulu-No-Lwazi
Automatic Speech Recognition • 0.3B • Updated • 18 -
uctnlp/omniASR-CTC-300M-v2-Zulu-All-v1
Automatic Speech Recognition • 0.3B • Updated • 12
models 10
uctnlp/omniASR-CTC-3B-v2-Zulu-All
Automatic Speech Recognition • 3B • Updated • 35
uctnlp/omniASR-CTC-300m-v2-Zulu-Baseline
Automatic Speech Recognition • 0.3B • Updated • 33
uctnlp/omniASR-CTC-300M-v2-Zulu-RDM-ANV
Automatic Speech Recognition • 0.3B • Updated • 22
uctnlp/omniASR-CTC-300M-v2-Zulu-No-Lwazi
Automatic Speech Recognition • 0.3B • Updated • 18
uctnlp/omniASR-CTC-300M-v2-Zulu-Lwazi
Automatic Speech Recognition • 0.3B • Updated • 15
uctnlp/omniASR-CTC-300M-v2-Zulu-All-v2
Automatic Speech Recognition • 0.3B • Updated • 69
uctnlp/omniASR-CTC-300M-v2-Zulu-All-v1
Automatic Speech Recognition • 0.3B • Updated • 12
uctnlp/omniASR-CTC-300M-v2-Zulu-AUG-ANV
Automatic Speech Recognition • 0.3B • Updated • 17
uctnlp/omniASR-CTC-1B-v2-Zulu-All
Automatic Speech Recognition • 1.0B • Updated • 33
uctnlp/mzansilm-125m
Text Generation • 0.1B • Updated • 1.54k • 32
datasets 6
uctnlp/AfriBED
Viewer • Updated • 12.8k • 145
uctnlp/mzansi-text
Viewer • Updated • 3.98M • 598 • 12
uctnlp/runyaner
Viewer • Updated • 267k • 67 • 1
uctnlp/t2x
Viewer • Updated • 4.7k • 47
uctnlp/mzansi-text-deduplicated
Viewer • Updated • 3.78M • 527 • 1
uctnlp/mzansi-text-tokenized
Viewer • Updated • 3.98M • 413