Amina Mardiyyah Rufai
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Research Interests
Amina Mardiyyah Rufai has a total of 6 co-authors. Here are the top co-authors by number of shared publications:
- Atnafu Lambebo Tonja
Instituto Politécnico Nacional 10 Technical University of Munich
1 shared publication - Bonaventure F. P. Dossou
McGill University
1 shared publication - Chris Chinenye Emezue
Mila Quebec AI Institute
1 shared publication - Sahib Singh
Ford Motor Company
1 shared publication - Tejumade Afonja
CISPA Helmholtz Center for Information Security
1 shared publication - Tobi Olatunji
Georgia Institute of Technology
1 shared publication
Latest Publications
- Open Access
AfriNames: Most ASR models "butcher" African Names
Useful conversational agents must accurately capture named entities to minimize error for downstream tasks, for example, asking a voice assistant to play a track from a certain artist, initiating navigation to a specific location, or documenting a laboratory result for a patient. However, where named entities such as ``Ukachukwu`` (Igbo), ``Lakicia`` (Swahili), or ``Ingabire`` (Rwandan) are spoken, automatic speech recognition (ASR) models' performance degrades significantly, propagating errors to downstream systems. We model this problem as a distribution shift and demonstrate that such model bias can be mitigated through multilingual pre-training, intelligent data augmentation strategies to increase the representation of African-named entities, and fine-tuning multilingual ASR models on multiple African accents. The resulting fine-tuned models show an 81.5\% relative WER improvement compared with the baseline on samples with African-named entities.
Open AccessAfriNames: Most ASR models "butcher" African Names
Useful conversational agents must accurately capture named entities to minimize error for downstream tasks, for example, asking a voice assistant to play a track from a certain artist, initiating navigation to a specific location, or documenting a laboratory result for a patient. However, where named entities such as ``Ukachukwu`` (Igbo), ``Lakicia`` (Swahili), or ``Ingabire`` (Rwandan) are spoken, automatic speech recognition (ASR) models' performance degrades significantly, propagating errors to downstream systems. We model this problem as a distribution shift and demonstrate that such model bias can be mitigated through multilingual pre-training, intelligent data augmentation strategies to increase the representation of African-named entities, and fine-tuning multilingual ASR models on multiple African accents. The resulting fine-tuned models show an 81.5\% relative WER improvement compared with the baseline on samples with African-named entities.