The Repository @ St. Cloud State

Open Access Knowledge and Scholarship



This paper provides basic information on developing speech synthesis to help preserve and revitalize critically endangered languages. It uses Betine, (ISO 639-3:eot), as a model. The death of minority languages is escalating globally, and 90% are predicted to die by 2100, which results in the loss of cultural heritage and knowledge. Synthesizing speech in dying and near-extinct languages can preserve and even revitalize them. The paper includes spectrographs and waveforms of the given Beti name transcribed phonetically as [aːɟo]. It also contains voice component measurements and the synthesizing programs used for comparison. The measurements taken from the data include F0/pitch, formants (F1, F2, F3, F4, F5), amplitudes (A1, A2, A3, A4), and A5, intensity, duration, and bandwidths (B1, B2, B3, B4, and B5). The paper notes that chunking longer phonemes into multiple short samples can give better synthesized results. Klatt, KlattGrid, and WORLD synthesizers are used for synthesizing the Betine speech segment. The paper proposes that vocoder synthesizers such as WORLD may be worth greater research concerning their capabilities in artificial speech synthesis.



To view the content in your browser, please download Adobe Reader or, alternately,
you may Download the file to your hard drive.

NOTE: The latest versions of Adobe Reader do not support viewing PDF files within Firefox on Mac OS and if you are using a modern (Intel) Mac, there is no official plugin for viewing PDF files within the browser window.