Kreitewolf Jens, Gaudrain Etienne, von Kriegstein Katharina
Max Planck Institute for Human Cognitive and Brain Sciences, Max Planck Research Group Neural Mechanisms of Human Communication, D-04103 Leipzig, Germany.
University of Groningen, University Medical Center Groningen, Department of Otorhinolaryngology/Head and Neck Surgery, 9700 RB Groningen, Netherlands; University of Groningen, Graduate School of Medical Sciences, Research School of Behavioural and Cognitive Neurosciences, 9713 GZ Groningen, Netherlands.
Neuroimage. 2014 May 1;91:375-85. doi: 10.1016/j.neuroimage.2014.01.005. Epub 2014 Jan 13.
Understanding speech from different speakers is a sophisticated process, particularly because the same acoustic parameters convey important information about both the speech message and the person speaking. How the human brain accomplishes speech recognition under such conditions is unknown. One view is that speaker information is discarded at early processing stages and not used for understanding the speech message. An alternative view is that speaker information is exploited to improve speech recognition. Consistent with the latter view, previous research identified functional interactions between the left- and the right-hemispheric superior temporal sulcus/gyrus, which process speech- and speaker-specific vocal tract parameters, respectively. Vocal tract parameters are one of the two major acoustic features that determine both speaker identity and speech message (phonemes). Here, using functional magnetic resonance imaging (fMRI), we show that a similar interaction exists for glottal fold parameters between the left and right Heschl's gyri. Glottal fold parameters are the other main acoustic feature that determines speaker identity and speech message (linguistic prosody). The findings suggest that interactions between left- and right-hemispheric areas are specific to the processing of different acoustic features of speech and speaker, and that they represent a general neural mechanism when understanding speech from different speakers.
理解来自不同说话者的语音是一个复杂的过程，特别是因为相同的声学参数既传达了有关语音信息的重要信息，也传达了说话者的相关信息。人类大脑在这种情况下如何完成语音识别尚不清楚。一种观点认为，说话者信息在早期处理阶段就被丢弃，不用于理解语音信息。另一种观点则认为，说话者信息被用来提高语音识别能力。与后一种观点一致，先前的研究发现了左右半球颞上沟/颞上回之间的功能相互作用，它们分别处理语音和说话者特定的声道参数。声道参数是决定说话者身份和语音信息（音素）的两个主要声学特征之一。在这里，我们使用功能磁共振成像（fMRI）表明，左右颞横回之间对于声襞参数也存在类似的相互作用。声襞参数是决定说话者身份和语音信息（语言韵律）的另一个主要声学特征。这些发现表明，左右半球区域之间的相互作用特定于语音和说话者不同声学特征的处理，并且它们代表了理解来自不同说话者语音时的一种普遍神经机制。
