More than 40% of the world’s 6000 languages are endangered. This NSF-funded project explores the use of speech signal processing and natural language processing, with a focus on deep learning automatic speech recognition (ASR), to document and preserve Seneca, an endangered language indigenous to what is now western New York State and Ontario. Together with members of the Haudenosaunee communities, our team of faculty and students in linguistics and computer science is collecting data and developing technologies that can be used by any community engaged in documenting a low-resource language.
For more information, explore the links above, visit the project's page at NSF.gov, or contact Dr. Emily Prud'hommeaux (emilypx@rit.edu).