Projects

Audiphon (Auditory models for automatic prediction of phonation)

Audiphon aims to build the foundations of the next generation of Automatic Phonation Recognition (APR)-systems. Using machine-learning techniques on features extracted from models that represent speech at different levels in the auditory system, Audiphon will provide phonation recognition methodologies with superior accuracy relative to those currently used in the study of under-resourced languages. Outcomes of Audiphon include software tools for APR and identification of auditory representations that best capture different phonation types (creaky, modal, breathy, etc.).

Microvariation in Bantu languages of South Africa: Building theories from typology data

This 6-year international project investigates the inter- and intra-language variation in Bantu languages as a collaboration between researchers in Japan and South Africa: the University of Venda in Limpopo and the University of the Free State. Conducting this project in these universities is ideal places because of the proximity to areas where speakers of Bantu languages are accessible. Superseding the 2-week workshop in March 2020 at the University of Venda, the current project aims to document a rich array of typological data and inform linguistic theories in a meaningful way. Specifically, this project has four specific goals:

1. Documenting inter- and intra-language variations at the interface between the prosody and sentence structures.

2. Documenting narratives for the discourse and dialect level variations.

3. Collecting archive-quality materials based on data obtained in (1) and (2).

4. Updating existing theories or proposing theories based on a new set of data.

This collaborative project will not only foster and cement research activities between Japan and South Africa, but also it will serve as a stepping stone for young researchers to engage in a long-term multi-faceted research program. Ultimately, the project will deepen our understanding of how the prosodic component and the morphosyntactic component interact in human languages.

A crosslinguistic study of prosody of particles: Japanese and Bantu languages

Co-PI: Toshio Matsuura (Hokusei Gakuin)

            Daisuke Shinagawa (TUFS, AA-ken)

Overview

This 3-year project investigates what role functional words such as particles play in the prosodic grammar. Japanese is a language that uses pitch to distinguish meaning between words; the word [ámé] with two high-pitched syllables means ‘candy’, whereas [àmê] pronounced with a low pitch followed by a high pitch means ‘rain'. Functional words are often short (less than three syllables), and prosodic effects of these words are yet to be fully understood. Building upon previous studies, this project focuses on prosodic effects of functional words in nominal as well as verbal domains in four types of languages based on the presence of tone and the position of particles: 

(a) Standard Japanese with pitch accent and suffixal particles 

(b) Aizu Japanese without pitch accent and suffixal particles

(c) Xitsonga with tone and prefixal particles

(d) Swahili without tone and prefixal particles

多言語による日本語学用語辞典および日琉諸語の用例に対するグロス規範の作成

PSYPHON: Psychoacoustic features for Phonation prediction

Co-PI: Konstantin Markov (Aizu)

Psychoacoustic features for phonation prediction (psyphon) aims at creating psychoacoustic based predictors of phonation by means of Machine Learning (ml) techniques. One of the main outcomes at the end of psyphon is a software tool to predict various types of phonation (e.g., modal, breathy, creaky, etc.) based on psychoacoustic models (roughness, loudness, etc.). Other outcomes are the creation of psychoacoustic models using ml, identification of psychoacoustic features related to different phonations, advancing our knowledge of the role of phonation in several languages, and helping to bridge the gap between under- and sufficiently-resourced languages.

A non-invasive test for diagnosing cognition in three SOV languages: Japanese, Korean and Bodo  

This project proposes that a non-invasive cognitive test focusing on sentence processing be developed for assisting elder population in societies such as Japan where the share of elder population is sharply increasing. More specifically, this project will make use of a series of cognitive-linguistic tests (Korean Patent #10-1562105) developed by Prof. Jee Eun Sung at Ewha Womans University in Korea. Three languages (Japanese, Korean and Bodo) will serve as core cases for gathering data from younger and elder population. Our findings based on solid linguistic theories will complement findings from other research programs in gerontology, psychology and medicine. Results could also reveal whether the onset of any change in cognitive abilities can be detected early on; a significant finding for societies with increasing elder population.

Establishment of a Research Network for Exploring the Linguistic Diversity and Linguistic Dynamism in Africa 

• [ReNeLDA homepage]

Africa is a continent embracing one third of world’s languages. Its linguistic diversity is not solely defined by the existence of various indigenous languages, a number of which have not been linguistically described, also characterized by people's daily practices of multilingual communication that bring ever changing dynamism to this linguistically and culturally diverse continent.  Based on this understanding, we aim to explore the linguistic diversity and dynamism of Africa by establishing an inter-institutional research network involving leading institutions working for linguistic description and documentation of African indigenous languages based in Africa and in Japan, with the following main objectives. 

Korean at the Nexus of the Northeast Asian Linguistic Area

"The project, “Korean at the Nexus of the Northeast Asian Linguistic Area,” will examine the status of Northeast Asia as a linguistic area or sprachbund. Korean shares features with languages spoken to the north, east, and south, reflecting the central cultural and political placement of Korea that continues to this day. The project team is comprised of ten researchers from Asia, Europe, and North America, including experts in both the diachronic and synchronic linguistics of Northeast Asia. The team will grow to include postdoctoral and graduate assistant researchers. International conferences are planned in Seoul in 2018 and St. Petersburg, Russia, in 2020, in addition to several workshops at Cornell." 

(this excerpt is taken from here)

Aging-related decline in sentence processing from crosslinguistic perspectives: Comparisons of SVO (English and SOV (Korean and Japanese) languages)

Purposes of the current project are 1) to investigate the effects of syntactic and semantic manipulations on aging-related decline in sentence processing and 2) to examine these effects from crosslinguistic perspectives by comparing the SOV (Subject-Object-Verb) languages (Korean and Japanese) to SVO (Subject-Verb-Object) language (English). Specific aims are as below: 1) To examine the effects of syntactic manipulation on sentence processing by manipulating the placement of prepositional (e.g., English) and postpositional phrases (e.g., Korean and Japanese, 2) To investigate the effects of semantic felicity manipulations on sentence processing by varying the degree of semantic felicity (felicitous vs. infelicitous) in prepositional and postpositional phrases, and 3) To identify whether there are differential effects of syntactic and semantic manipulations on age-related changes across the languages.

Phonetics Phonology and New Orthographies: Helping Native Language Communities in the Himalayas (PhoPhoNo)

• [Project website] [Official grant information]

The research project aims to conduct phonological research on three Trans-Himalayan languages, namely the Dränjoke language of Sikkim and two Tamang lects, and to develop orthographies for each of these languages. PhoPhoNO represents a joint project between the University of Bern in Switzerland and International Christian University of Japan and will embody a new approach to the documentation and description of linguistic structures, by combining the expertise and competence of scholars working in different domains and allowing more in-depth research. The linguists in Bern and in Tokyo will use their expertise to help three native language communities in the Himalayas.

Phonetic typology of consonant-tone interaction

This project investigates acoustic, articulatory and auditory phonetic data of consonants in Xitsonga, SiSwati and Japanese. Xitsonga and SiSwati have consonants that lower pitch, called depressors. In addition, these consonants interact with tonal processes such as high tone spreading. Even so, there has been no systematic phonetic study of these consonants. This project aims to fill the gap. Moreover, Korean and Tshivenda data has been collected as part of the project. 

Data collection will occur in Thohoyandou, South Africa and Tokyo, Japan. Articulatory data will be collected using an Electroglottograph.