首页> 外文期刊>日本音響学会誌/The Journal of the Acoustical Society of Japan >Abstracts of Papers in Acoustical Science and Technology
【24h】

Abstracts of Papers in Acoustical Science and Technology

机译:声学科学与技术论文的摘要

获取原文
获取原文并翻译 | 示例
           

摘要

Aiming at scientific understanding of color association after listening to vowel sounds, cross-modal association characteristics was investigated using parameters directly extracted from speech and color. A perceptual experiment was conducted by employing five Japanese vowel sounds produced by multiple male and female native speakers differing in spectral characteristics. For color description, three attributes (hue, saturation and value) of 153 color tips of Practical Color Coordinate System (PCCS) selected after listening to vowel speech were employed. To enable scientific interpretation, acoustic features of vowel sounds (F0, pitch range, intensity, F1, F2, F3, F4) were used. Regression analyses were carried out between these speech and color parameters. From the results using multiple linear regression and neural networks, significant correlations could have observed between F1, pitch range & F2 and hue, F0 & F2 and value of color. These experimental findings can exactly and clearly show vowel-color association characteristics which have been partially studied in phonetics and cognitive science fields mainly using speech and color categories. These precise parametric correlations between color and speech can provide scientific knowledge for further investigations of cross-modal correlations in multiple research fields which will give the new possibilities for multi-modal information expression.
机译:旨在科学了解聆听元音后的颜色关联,使用直接从语音和颜色中提取的参数来研究跨模型关联特征。通过使用多个男性和女性母语扬声器产生的五个日本元音声音来进行感知实验。对于颜色描述,采用了在收听元音语音后选择的实用颜色坐标系(PCCS)的三个属性(色调,饱和度和值)。为了实现科学解释,使用元音声音的声学特征(F0,俯仰范围,强度,F1,F2,F3,F4)。在这些语音和颜色参数之间进行回归分析。根据使用多元线性回归和神经网络的结果,F1,间距和F2和色调,F0&F2和颜色值之间可以观察到显着的相关性。这些实验结果可以完全清楚地显示在语音和认知科学领域部分地研究的元音结合特征,主要使用语音和彩色类别。颜色和语音之间的这些精确的参数相关性可以提供科学知识,以进一步调查多个研究领域中的跨模型相关性,这将为多模态信息表达提供新的可能性。

著录项

相似文献

  • 外文文献
  • 中文文献
  • 专利
获取原文

客服邮箱:kefu@zhangqiaokeyan.com

京公网安备:11010802029741号 ICP备案号:京ICP备15016152号-6 六维联合信息科技 (北京) 有限公司©版权所有
  • 客服微信

  • 服务号