A Correlational and Predictive Analysis of the Relationship between F0 and the Formants of Romanian Vowels

被引:0
|
作者
Gheltu, Stefan-Andrei [1 ]
机构
[1] Gheorghe Asachi Tech Univ Iasi, Fac Elect Telecommun & Informat Technol, Iasi, Romania
关键词
formants; pitch; vowels; emotions; fluent speech; regression; Romanian language;
D O I
10.1109/ecai50035.2020.9223249
中图分类号
TP18 [人工智能理论];
学科分类号
081104 ; 0812 ; 0835 ; 1405 ;
摘要
We are interested in investigating the manner in which emotions influence the level of correlation of the formants of Romanian vowels during dynamic pronunciation. In this study we analyzed the statement /vine mama/ (mother is coming) from which we extracted for each vowel the fundamental frequency and the formants, and we studied the variations of the correlation coefficients calculated for F0 and F1-F4 according to the emotion expressed by the speaker (joy, sadness, anger and the state of emotional neutrality). Subsequently, we applied simple linear regression to verify whether the pitch is an effective predictor for estimating the low formants (F1 and F2) or not. Some of the results we obtained indicate that for both /a1/ and /a2/ the values of the coefficient of determination (R2) were low regardless of the type of emotion. As for vowel /a1/, the pitch is a good predictor of formant F1 for anger and sadness, and for F2 it is an effective predictor only in the case of anger and neutral tone. On the other hand, for vowel /a2/, F0 is a good predictor for both F1 and F2, but only in association with anger.
引用
收藏
页数:6
相关论文
共 50 条
  • [31] ARTICULATORY HYOID-LARYNGEAL CHANGES VS CRICOTHYROID MUSCLE-ACTIVITY IN THE CONTROL OF INTRINSIC F0 OF VOWELS
    VILKMAN, E
    AALTONEN, O
    RAIMO, I
    ARAJARVI, P
    OKSANEN, H
    JOURNAL OF PHONETICS, 1989, 17 (03) : 193 - 203
  • [32] Introducing the f0% method: a reliable and accurate approach for qPCR analysis
    Gamal, Mahmoud
    Ibrahim, Marwa A.
    BMC BIOINFORMATICS, 2024, 25 (01)
  • [33] F0 estimation of noisy speech based on complex speech analysis
    Kinjo, Tatsuhiko
    Funaki, Keiichi
    2006 IEEE 12TH DIGITAL SIGNAL PROCESSING WORKSHOP & 4TH IEEE SIGNAL PROCESSING EDUCATION WORKSHOP, VOLS 1 AND 2, 2006, : 434 - 437
  • [34] Variable f0 of toneless moras in Suzhou Chinese: a computational analysis
    Zhu, Yuhong
    GLOSSA-A JOURNAL OF GENERAL LINGUISTICS, 2024, 9 (01):
  • [35] Introducing the f0% method: a reliable and accurate approach for qPCR analysis
    Mahmoud Gamal
    Marwa A. Ibrahim
    BMC Bioinformatics, 25
  • [36] F0 analysis of perceptual distance among Cantonese level tones
    Ho, Rerrario Shui-Ching
    Sagisaka, Yoshinori
    INTERSPEECH 2007: 8TH ANNUAL CONFERENCE OF THE INTERNATIONAL SPEECH COMMUNICATION ASSOCIATION, VOLS 1-4, 2007, : 2756 - 2759
  • [37] F0, LPC, and MFCC Analysis for Emotion Recognition Based on Speech
    Teixeira, Felipe L.
    Teixeira, Joao Paulo
    Soares, Salviano F. P.
    Pio Abreu, J. L.
    OPTIMIZATION, LEARNING ALGORITHMS AND APPLICATIONS, OL2A 2022, 2022, 1754 : 389 - 404
  • [38] F0-dependence, number alteration, and non-systematic behaviour of the formants in German vowels (vol 83, pg 25, 1995)
    Maurer, D
    Landis, T
    INTERNATIONAL JOURNAL OF NEUROSCIENCE, 1996, 86 (1-2) : 167 - 167
  • [39] ANALYSIS AND SYNTHESIS OF GERMAN F0 CONTOURS BY MEANS OF FUJISAKI MODEL
    MOBIUS, B
    PATZOLD, M
    HESS, W
    SPEECH COMMUNICATION, 1993, 13 (1-2) : 53 - 61
  • [40] The relationship of vocal loudness manipulation to prosodic F0 and durational variables in healthy adults
    Watson, Peter J.
    Hughes, Deanna
    JOURNAL OF SPEECH LANGUAGE AND HEARING RESEARCH, 2006, 49 (03): : 636 - 644