Research Overview
Our research spans multiple areas of language technology, from speech recognition to natural language processing and machine learning. Here is an overview of our current research focus areas.
Speech Technology
Developing automatic speech recognition, text-to-speech synthesis, and speaker identification systems for Yoruba language and its dialects.
Current Projects
- Building ASR models for Yoruba with support for multiple dialects
- Developing natural-sounding TTS systems using neural networks
- Creating speaker identification systems for forensic and security applications
Key Datasets
Speech Recognition Dataset (50K+ hours)TTS Voice Dataset
Natural Language Processing
Building NLP tools for Yoruba dialects including machine translation, sentiment analysis, and named entity recognition.
Current Projects
- Developing Yoruba-English machine translation systems
- Creating named entity recognition systems for information extraction
Key Datasets
Text Corpus (2M+ samples)Parallel Translation (500K+ pairs)
Machine Learning
Developing models and frameworks specifically designed for Yoruba dialects and low-resource languages.
Current Projects
- Adapting multilingual models for Yoruba using transfer learning
- Optimizing models for resource-constrained environments
- Developing few-shot learning techniques for low-resource scenarios
Publications & Papers
Check back soon for our latest publications, or contact us to learn more about our research.