Mono-pitched expression of emotions in different vowels
Teija Waaramaa , Anne-Maria Laukkanen, Paavo Alku, Eero Väyrynen
Fundamental frequency (F0) and intensity are known to be important variables in communication of emotions in speech. In singing, however, the pitch is predetermined and yet the voice should convey emotions. The present study investigated the role of voice source characteristicsand formant frequenciesin communication of emotions in mono-pitched samples. Student actors (5 males, 8 females) produced samples on vowels [a:], [i:] and [u:] simulating joy, tenderness, sadness, anger and a neutral emotional state. The samples were analyzed for equivalent sound level (Leq), duration, alpha ratio and formant frequencies F1-F4. The [a:] samples were inverse filtered and the estimated glottal flows were parameterized with NAQ (normalized amplitude quotient). Forty participants listened to the randomized samples (N= 210) for identification of the emotions. Analysis of variance (ANOVA) was used to study the parameter interrelations. The samples were recognized with 50 % accuracy (intrarater reliability 59 %). Vowel [a:] conveyed best tenderness, [i:] joy and [u:] sadness and neutrality. Anger was carried well by all of the vowels. Leq, NAQ and F2 and F3 were discriminating parameters in both genders. When Leqwas set as a covariate, alpha ratio and F1, F2 and F4 appeared to remain significant in some vowels. The results suggest that when pitch was predetermined, the main variable in vocal expression of emotions was Leq. Additionally, phonation type and articulation (alpha ratio and formant frequencies) tended to be used in the expression of emotions, to some extent independently of Leq.
Keywords: voice quality, inverse filtering, voice source, formants, mono-pitched vowels, emotions, perception.
Teija Waaramaa¹, , Anne-Maria Laukkanen¹, Paavo Alku2, Eero Väyrynen3
¹Department of Speech Communication and Voice Research, University of Tampere, Tampere, Finland
2MediaTeam, University of Oulu, Oulu, Finland
3Laboratory of Acoustics and Audio Signal Processing, Helsinki University of Technology, Espoo, Finland