A new paper by Zhihao and Youngah, Tone contour identification from the perspective of production, has been published in the Journal of East Asian Linguistics.
The study addresses a recurring challenge in tone research: the same linguistic tone can sometimes be described as having different contour shapes across different sources. One reason is that a tone’s intended contour can be distorted when it is actually produced, making it difficult to determine whether its target shape is level, rising, or falling from the observed pitch pattern alone.
The results show that commonly used measurements such as pitch range and slope can sometimes be ambiguous, particularly when distinguishing level tones from rising or falling tones with relatively small pitch ranges. The study instead identifies three measures related to the “steadiness” of pitch movement that provide a clearer distinction between level and non-level contours. A case study applying the method to tones from the Xiangtan and Jian’ou Chinese dialects further demonstrates how these measures can help resolve ambiguous contour classifications.
The study provides a broader set of quantitative diagnostics for identifying tone contours from production data and offers a new perspective on why contour transcriptions may differ across descriptions of Chinese dialects. Future work can extend this production-based approach to more complex contours and potentially contribute to automatic classification of linguistic tone contours.
Wang, Z., & Do, Y. (2026). Tone contour identification from the perspective of production. Journal of East Asian Linguistics, 35, Article 22. open_in_new DOI