Advancing linguistic research through speech technologies

This project addresses critical languages for research and technology development. By leveraging advanced speech technologies, this initiative bridges the gap between linguistic research and practical application. The project develops tools including automatic speech recognition and text-to-speech systems and creates educational resources for language learners and researchers. Through collaboration with communities, the project ensures that these technologies are innovative, effective, and appropriate. Other benefits to society include a demonstration of the transformative potential of combining technology with language research. This project addresses the challenge of limited linguistic resources, which hinders the development of effective natural language processing tools and algorithms. By fine-tuning pre-existing speech technologies and applying transfer learning, the project creates accurate and scalable solutions for language documentation and analysis. Key activities include the development of benchmark datasets, the creation of comprehensive lexica, and the implementation of advanced data processing techniques such as tokenization, lemmatization, and part-of-speech tagging. These resources are made publicly available, enabling broader research and educational use. Through its interdisciplinary approach, combining computational linguistics with language documentation, the project advances linguistic research and promotes digital literacy. This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.