Evaluating phonemic transcription of low-resource tonal languages for language documentation

TitleEvaluating phonemic transcription of low-resource tonal languages for language documentation
Publication TypeConference Paper
Year of Publication2018
AuthorsAdams O, Cohn T, Neubig G, Cruz H, Bird S, Michaud A
Published inProceedings of the LREC 2018 Workshop
KeywordsAsian languages, language documentation, low-resource languages, Mesoamerican languages, speech recognition
Abstract

Transcribing speech is an important part of language documentation, yet speech recognition technology has not been widely harnessed to aid linguists. We explore the use of a neural network architecture with the connectionist temporal classification loss function for phonemic and tonal transcription in a language documentation setting. In this framework, we explore jointly modelling phonemes and tones versus modelling them separately, and assess the importance of pitch information versus phonemic context for tonal prediction. Experiments on two tonal languages, Yongning Na and Eastern Chatino, show the changes in recognition performance as training data is scaled from 10 minutes up to 50 minutes for Chatino, and up to 224 minutes for Na. We discuss the findings from incorporating this technology into the linguistic workflow for documenting Yongning Na, which show the method’s promise in improving efficiency, minimizing typographical errors, and maintaining the transcription’s faithfulness to the acoustic signal, while highlighting phonetic and phonemic facts for linguistic consideration. 

Acknowledgment

We are very grateful for support from NSF Award 1464553 Language Induction Meets Language Documentation. 

URLhttps://halshs.archives-ouvertes.fr/halshs-01709648v4/document
ICSI Research Group

AI