Chan Karen M K, Yiu Edwin M-L
Voice Research Laboratory, Department of Speech and Hearing Sciences, The University of Hong Kong.
J Speech Lang Hear Res. 2002 Feb;45(1):111-26. doi: 10.1044/1092-4388(2002/009).
Perceptual voice evaluation is a common clinical tool for rating the severity of vocal quality impairment. However, the evaluation process involves subjective judgment, and reliability is therefore a major issue that needs to be considered. When listeners are asked to judge the quality of a voice signal, they use their own internal standards as the references. These internal standards can be variable, as different individuals may have acquired different standards in prior situations. In order to improve the reliability of the perceptual voice evaluation process, external anchors and training are provided to counteract the effect of these internal standards. This study investigated to what extent the provision of anchors and a training program would improve the reliability of perceptual voice evaluation by naive listeners. The results show, in general, that anchors and training helped to improve the reliability of perceptual voice evaluation, especially in the rating of male voices. Furthermore, it was found that anchors made up of synthesized signals combined with training were more effective in improving reliability in judging perceptual roughness and breathiness than natural voice anchors.
感知性嗓音评估是一种用于评定嗓音质量损害严重程度的常用临床工具。然而，评估过程涉及主观判断，因此可靠性是一个需要考虑的主要问题。当要求听众判断语音信号的质量时，他们会以自己的内部标准作为参考。这些内部标准可能会有所不同，因为不同的个体在先前的情况下可能已经获得了不同的标准。为了提高感知性嗓音评估过程的可靠性，提供了外部锚点和培训，以抵消这些内部标准的影响。本研究调查了提供锚点和培训计划在多大程度上会提高新手听众感知性嗓音评估的可靠性。结果总体表明，锚点和培训有助于提高感知性嗓音评估的可靠性，尤其是在男性嗓音的评分方面。此外，还发现由合成信号组成的锚点与培训相结合，在提高判断感知粗糙度和呼吸声的可靠性方面比自然嗓音锚点更有效。
