| 研究課題/領域番号 |
20K11956
|
|---|
| 研究機関 | 会津大学 |
|---|
研究代表者 |
Julian Villegas 会津大学, コンピュータ理工学部, 上級准教授 (50706281)
|
|---|
| 研究分担者 |
李 勝勲 国際基督教大学, 教養学部, 上級准教授 (20770134)
MARKOV K 会津大学, コンピュータ理工学部, 教授 (80394998)
|
|---|
| 研究期間 (年度) |
2020-04-01 – 2023-03-31
|
| キーワード | Corpus acquisition / Machine learning |
|---|
| 研究実績の概要 |
We performed field recordings of Zapotec and Mixe languages in Oaxaca (Mexico). We also conducted several experiments to fine-tune the hyperparameters of the machine learning algorithms. Finally, we conducted an extensive literature review, tabulating the results of twelve of studies for the prediction models. In addition, we conducted several subjective experiments mainly on the effect of linguistic background on the classification of creaky words.
|
| 現在までの達成度 (区分) |
現在までの達成度 (区分)
3: やや遅れている
理由
The effort of the three groups and their periodic communication have been reflected in the smooth progress of the project, arriving at the third year a bit behind schedule after the hiatus in data collection resulting from the COVID-19 pandemic countermeasures. We have increased our efforts to finish the activities related to the post-processing of the collected data to bring the schedule back on track.
The results of the subjective experiments have been analyzed already and in preparation for publication. Likewise, the literature review on Zapotec languages has been published already in [1].
[1] Y. Baldoria, B. P. Fleming, M. W. R. Liu, J. Villegas, and S. J. Lee, ICU language database series 8: Comparative Zapotec,” in ICU Working Papers in Linguistics, no. 22, pp. 225-349.
|
| 今後の研究の推進方策 |
We plan to finish the development related to machine learning this year. Simultaneously, we will focus on publishing hitherto results, as contemplated in our original plan. Since we have more momentum on the analysis of creakiness and roughness, we will proceed first with these, and later with psychoacoustic sharpness and other kinds of phonation such as breathy voice. Towards the end of the year, we will release several tools including one for the manual evaluation of phonation. This software application is currently being used for assessing the phonation of words by expert judges.
|
