Matthew R. Lee
Language Technology Consultant at SIL International

The transcription bottleneck (Seifart et al. 2018, 1) delays and sometimes prohibits valuable language documentation resources from being shared with the target community and other researchers. An oral-first annotation methodology (Boerger et al. 2019) partially alleviates that problem through a process of oral annotation that can be done much more efficiently than written transcription. Specialized tools such as SIL International's SayMore aid a researcher and community to facilitate collection, annotation, documentation, and archiving of orally and textually annotated data in a standardized format, but the resulting web of aligned archival data is disjointed and far from an interactive format ready to be experienced by the community. This thesis proposes a solution, a tool named Prestige, that complements oral annotation's contribution by providing an automatically generated multilayered media experience designed to mobilize the resulting rich sources and annotations for immediate use in the community.

Despite recent advances in natural language processing and other language technology, the application of such technology to language documentation and conservation has been limited. In August 2019, a workshop was held at Carnegie Mellon University in Pittsburgh to attempt to bring together language community members, documentary linguists, and technologists to discuss how to bridge this gap and create prototypes of novel and practical language revitalization technologies. This paper reports the results of this workshop, including issues discussed, and various conceived and implemented technologies for nine languages: Arapaho, Cayuga, Inuktitut, Irish Gaelic, Kidaw'ida, Kwak'wala, Ojibwe, San Juan Quiahije Chatino, and Seneca.

Multilingual models can improve language processing, particularly for low resource situations, by sharing parameters across languages. Multilingual acoustic models, however, generally ignore the difference between phonemes (sounds that can support lexical contrasts in a particular language) and their corresponding phones (the sounds that are actually spoken, which are language independent). This can lead to performance degradation when combining a variety of training languages, as identically annotated phonemes can actually correspond to several different underlying phonetic realizations. In this work, we propose a joint model of both language-independent phone and language-dependent phoneme distributions. In multilingual ASR experiments over 11 languages, we find that this model improves testing performance by 2% phoneme error rate absolute in low-resource conditions. Additionally, because we are explicitly modeling language-independent phones, we can build a (nearly-)universal phone recognizer that, when combined with the PHOIBLE [1] large, manually curated database of phone inventories, can be customized into 2,000 language dependent recognizers. Experiments on two low-resourced indigenous languages, Inuktitut and Tusom, show that our recognizer achieves phone accuracy improvements of more than 17%, moving a step closer to speech recognition for all languages in the world.

A great Scripture engagement opportunity lies before us. People are publicly reading the Scriptures every Sunday using lectionaries in the national language. Especially for minority languages without a full Bible, how can we make these important readings available in the mother tongue?

This presentation seeks to increase awareness about some of the Language Technology Tools that support work in the Languages in Cameroon.

Cette présentation a été presenté pour partager quelque connaissances sur l'archivage et le copyright aux étudiants de l'Université à une discussion table round.

This presentation presents a technical history of the Cameroon Keyboard, and makes some proposals for standardizing the technical specifics of the General Alphabet of Cameroonian Languages.

This overview presentation was given to SIL Cameroon's Advisory Board for their review of tools and Activities of CMB's Language Technology Department.

Now that dictionaries are no longer stored as formatted documents, but as lexical databases (in FLEx, WeSay, Toolbox, etc.), what can we do with this all this structured data beyond printing dictionaries and can we share this wealth? This project (under the guidance of Gary Simons) takes the smart data of FLEx lexical databases a step further. By linking together our lexicons stored in appropriate forms, we can start to leverage the structured knowledge within to ask interesting cross-linguistic questions of our data, and we can share that resource with the outside world.
• Do any of the languages in this cluster have a non-borrowed word for faith?
• Is the past tense in each language marked with a prefix, suffix, or clitic?
• In these related languages, are person and number typically expressed in separate morphemes or as a portmanteau?
Recorded on May 8, 2014 in Mahler 5-7 at GIAL.
Edited by Matthew. Property of Matthew Lee, Graduate Institute of Applied Linguistics, and SIL International

This is my capstone thesis for my Bachelor's in Integrated Science and Technology at James Madison University

This is the Blog of the Language Technology Department for SIL Cameroon, providing news and resources for the minority languages of Cameroon. Special attention is gieven to Keyboards, fonts, and encoding.

The Master of Arts degree with a major in Applied Linguistics is designed to produce graduates qualified to serve in specialist cross-cultural roles in Descriptive Linguistics.

After the groundwork in the Certificate in Applied Linguistics, this program continued with courses on Language Documentation, Advanced Phonology, Sonority, Syntax, Discourse, Linked Data in Linguistics, Semantics and Pragmatics, Cross-cultural Training, and related fields.

My chosen courses have provided a solid grounding in many domains of linguistics, with focus on available and emerging technologies that can aid such work.

• Develop French and English materials and train users in effective use of linguistic and translation software. (Paratext, FieldWorks, Adapt It, WeSay, Phonology Assistant, OurWord, etc.)
• Customize and troubleshoot linguistic applications for various languages and situations.
• Convert and clean large documents and data for publishing or import. (Regular Expressions, EditPad Pro, PowerGrep, MS Office, Libre/OpenOffice, Acrobat)
• Develop and maintain customized keyboards for multiple operating systems and languages.
• Advocate new and traditional technologies and workflows through networking and research.

• Instructor for the Language and Culture Documentation course

• Develop French and English materials and train users in effective use of linguistic and translation software. (Paratext, FieldWorks, Adapt It, WeSay, Phonology Assistant, OurWord, etc.)
• Customize and troubleshoot linguistic applications for various languages and situations.
• Convert and clean large documents and data for publishing or import. (Regular Expressions, EditPad Pro, PowerGrep, MS Office, Libre/OpenOffice, Acrobat)
• Develop and maintain customized keyboards for multiple operating systems and languages.
• Advocate new and traditional technologies and workflows through networking and research.

• Started course as the Teaching Assistant.
• Took over as instructor for 7 of 8 weeks of the grad-level technology course during the professor’s emergency leave.
• Taught FieldWorks Language Explorer, Phonology Assistant, Linguistic Desktop Publishing, LingTree, etc.

• Led six-week course on effective use of linguistic and translation software. (Paratext, Translator’s Workplace, WeSay, etc.)

• Managed remote monitoring system to provide maintenance and checks of customer’s critical systems.
• Telephone system installation, configuration, and troubleshooting. Focus on NEC Elite IPK II.
• Installed, analyzed, maintained, and supported both our office network and customer networks. (server/router/wireless configuration, switch/bridge/PC installs, replace devices, wiring)
• Created sales orders, order supplies, negotiate with customers, complete contractor paperwork

• Taught 2D & 3D Video Game Creation using different environments and variety of multimedia apps.
• Fulfilled full 24-hour camp counselor duties (leading activities, games, supervising dormitory)

• Troubleshoot computer and user issues, software training, computer setup
• Server maintenance, database design and management, data entry