<p>This study explores how timbral features shape perceived emotions in Chinese traditional instrumental music, offering insights into the interplay between music and culture. Using a dataset of 273 music excerpts annotated by 168 Chinese participants, we analyzed the relationships between 54 timbral features and perceived emotions, considering both the dimensional model of emotion (valence and arousal) and discrete emotions (anger, sadness, happiness, peacefulness, transcendence, gentleness, and solemnness). Linear regression analyses revealed that timbral features commonly used in Western music studies can also effectively predict the perceived emotions in Chinese traditional music, with adjusted R² values ranging from 0.302 (solemnness) to 0.751 (arousal). Combining linear regression and random forest models, we identified distinct emotional expression patterns unique to Chinese traditional music. Most notably, happiness was idiosyncratically associated with increased noise energy and inharmonicity, a pattern that contrasts with findings in many Western music studies. By contrast, sadness corresponded to reduced noise energy and lower inharmonicity, solemnness was characterized by a narrower range of spectral variability and shorter sound duration, and transcendence was expressed through reduced spectral variability, a wider range of spectral kurtosis, and a narrower range of spectral flatness. These findings may be closely connected to Chinese traditional culture, warranting further in-depth investigation.</p>