Exploring acoustic differences between Cantonese (tonal) and English (non-tonal) spoken expressions of emotions

It has been claimed that tone language speakers use less F0 related cues in the production of verbal expressions of emotions. This is because F0 is used in the production of lexical tones. This study investigated this claim by examining how F0 and various other acoustic parameters are used in the production of verbal emotion expressions in Cantonese (tone language) compared to English (non-tone language). Acoustic measurements (e.g., mean F0, F0 range) were extracted from the verbal expressions of five emotions (angry, happy, sad, surprise and disgust) and a neutral expression produced by five male native speakers of Cantonese and English. They were analyzed using Kmeans clustering to see how different acoustic properties are grouped and how this varies as a function of language. The results showed some difference between the two languages in how F0 related cues are used in the production of emotions. The results are discussed in terms of the general acoustic characteristics of spoken emotion expressions and in relation to behavioral data from perceptual studies.