Community resources
Mozilla Common Voice - Hakha Lai
In summer 2018, as a first step towards the eventual development of speech technology, we began building a Laiholh corpus using the Mozilla Common Voice platform. In our own words (Berkson et al. 2019):
The ultimate goal of our team is to contribute to the development of speech translation technology that will be of benefit, both in general and in the local community in Indianapolis. Translation tools would be of great use in local emergency rooms, schools, and businesses.
Our long-term goal is to develop automatic speech translation for Laiholh, and Mozilla’s Common Voice platform offers two necessary outcomes that bring us closer to that goal. It facilitates both the creation of a public domain spoken corpus and the development of speech-to-text software. Speech data is collected via a phone or browser app from any native speaker willing to donate their voice to the corpus. Once the corpus is large enough, Mozilla will use machine learning software to develop speech-recognition technology. This moves us closer to our larger goal of speech translation because once a written Laiholh sentence can be generated from spoken language, that written language can be used as input for text-based (Laiholh–English or English–Laiholh) machine translation technologies.
(access our article about Common Voice here)
Undergraduate research internships
Support from the IU Office of Undergraduate Research and the IU Social Science Research Commons funded three UG research internships in Summer 2019 which culminated in poster presentations at an IU-wide event (pictured below). Thomas Thawngza (mentor: James C. Wamsley) presented “A Survey on Hakha Lai Acquisition and Attitudes”; Zai Sung (mentor: Samson Lotven) presented “The Zophei Verbal Complex”; Sui Hnem Par (mentors Berkson, Lotven) presented “Literacy Efforts in Lutuv”.
