ISSN 1671-3710
CN 11-4766/R
主办:中国科学院心理研究所
出版:科学出版社

心理科学进展 ›› 2011, Vol. 19 ›› Issue (6): 809-813.

• 研究简报 • 上一篇    下一篇

听音识人—— 语音频谱与人格特质的关系初探

胡超;傅根跃   

  1. 浙江师范大学心理系, 金华 321004
  • 收稿日期:2011-04-18 修回日期:1900-01-01 出版日期:2011-06-15 发布日期:2011-06-15
  • 通讯作者: 傅根跃

“Perceive One’s Character through His Voice”: The Relationship between Speech Spectrum and Personality Traits

HU Chao;FU Gen-Yue   

  1. Department of Psychology, Zhejiang Normal University, Jinhua, Zhejiang 321004, China
  • Received:2011-04-18 Revised:1900-01-01 Online:2011-06-15 Published:2011-06-15
  • Contact: FU Gen-Yue

摘要: 本研究探索不同情绪状态下说话时单个音素的语音频谱与其人格特质的关系。45个被试参加了16PF人格特质测验, 然后对人格测评的虚假结果进行反馈, 同时报告自己的情绪感受。通过记录被试在不同情绪状态下的回答“是”和“不是”, 然后用语音软件Praat分析辅音/Sh/的频谱参数:前三个共振峰的频率和波段, 结果发现:在平均0.14秒时长的辅音/Sh/上就有了显著的情绪效应, 而且, 频谱参数从中性情绪状态到积极情绪状态的变化量与说话者的情绪稳定性、敢为性存在显著相关; 被试在不同情绪状态下音素/Sh/的频谱参数与部分人格特质有显著相关, 尤其是敢为性、敏感性这类与个体情绪、人际交往行为有密切关系的特质; 此外, 不同情绪状态和回答类型下语音频谱参数与人格特质相关不同。本研究结果预示了从语音频谱揭示说话者人格特质的可能性。

关键词: 语音频谱, 人格特质, 情绪状态

Abstract: The current research explored the correlation between speakers’ personality traits and speech spectrums of the same phoneme during different emotional states. 45 subjects completed a 16PF personality test, and then gave feedback on false results of personality test, describing their feelings at the same time. Their responses during different emotional states were recorded. Frequency and bandwidth of the former three formants of the consonant /Sh/ were analyzed by Praat voice software; results showed that the emotion’s effect is already significant on the consonant /Sh/, although its average duration is just 0.14 seconds. In addition, their variation from a neutral emotional state to a positive emotional state were significantly correlated with speakers’ Emotional Stability and Social Boldness. Speech spectrum parameters were significantly correlated with some of speakers’16PF personality traits during the associated emotional state, especially with Social Boldness and Sensitivity which are significantly correlated with individual’s emotional and interpersonal communication behaviors. Furthermore, the correlation varies between speakers’ different emotional states and response type. The results of this research indicated the possibility of detecting speakers’ personalities through their speech spectrums.

Key words: speech spectrum, personality traits, emotional state