The human vocal tract is the cavity in human beings where speech sounds are made as air flows from the lungs upward and travels through where the vocal cords are (the "larynx") and then the space above the laryngeal cavity (the "pharynx") until it exits either through the nasal cavity or through the mouth. Different parts ("articulators") modify the airflow to create speech sounds. Since many articulators are hidden from view when the speech sounds are produced, linguists have found other devices to investigate where and how articulators operate. This is important because different sounds can be meaningful in a particular language. For example, "p" and "b" are different sounds in English since "pit" and "bit" mean different things. Instrumental tools like ultrasound and devices that measure oral-nasal airflow enable the precise articulation of speech sounds to be understood. As a result, sounds will be more accurately transcribed in language documentation and produced by learners. Sound systems are also researched by collecting and analyzing the language used in more stylized ways, like rhyme or repeated consonant sounds ("alliteration") in different genres. Understanding the sound system of a language is critical for compiling dictionaries, developing a systematic orthographic system and language teaching materials. This project will document the speech sounds of a highly endangered language that has an unusually large number of sounds that are formed by curling the tip of the tongue up towards the hard palate ("retroflex"). Broader impacts include the production of linguistic and cultural materials for direct use by the endangered language community, the training of a post-doctoral researcher and undergraduate students, and a new international collaboration with a local university and an NGO. USAID researchers working on education and literacy projects in the region will also benefit from the products of this project. The NSF Office of International Science and Engineering is providing support for international activities associated with this project.
This project will bring together a team of linguists and local language documentation researchers to describe the sound system of Kalasha (3000-5000 speakers), an endangered Dardic language of Northern Pakistan. Kalasha is of linguistic importance in part due to its many retroflexed sounds, since the wide range of these and other speech sounds are found in less than 1% of the world's languages. The project will document the phonetics and phonology of Kalasha through instrumental techniques like lingual ultrasound, and will use best practices in language documentation to compile an online dictionary and to transcribe oral literature (songs, short stories, narratives, poems). The researchers will also develop flexible speech analysis tools, and make data and findings available through existing cross-linguistic databases and archives. Lingual ultrasound and oral-nasal airflow are ideal methods for describing the retroflex consonants, and nasal and retroflex vowels of Kalasha. The instrumental phonetic investigation will enable a better understanding of these uncommon speech sounds and will generate an audio-video ultrasound corpus. The audio-video corpus and other spontaneous speech data will be used to revise the orthography of Kalasha and to develop pedagogical materials for the community.
