Tracing the mechanisms that generate tonal content in voiced speech

追踪在有声语音中生成音调内容的机制

• 批准号: 446965891
• 负责人: Professor Dr.-Ing. Stefan Becker
• 金额: --
• 依托单位: Lehrstuhl für Strömungsmechanik (LSTM)
• 依托单位国家: 德国
• 项目类别: Research Grants
• 资助国家: 德国
• 项目状态: 未结题
• 来源: https://gepris.dfg.de/gepris/projekt/446965891?language=en
• 关键词: Tracing; mechanisms; generate; tonal; content

Voice disorders are of high social and economic relevance as it affects many people in our society. Consequently, concerned patients suffer from increased physiological and psychological stress as reduced quality of life.The normal voice of a healthy human is the result of a complex fluid-structure-acoustic interaction. The structural motion of the vocal folds intensely modulates the flow through the glottis, which generates the main portion of the sound. Thereby, normal and healthy phonation is characterized by a strong harmonic content. The vocal tract downstream of the vocal folds is a key factor for acoustic back-coupling on the vocal fold motion as well as the flow. Hence, in addition to creating formants, we hypothesize that the vocal tract also impacts the generation of higher harmonics in the resulting voice signal, which is crucial for a healthy voice. The central objective of the research project is to advance the understanding of tonal sound generation during phonation based on a fully-coupled experimental investigation of the underlying mechanisms. The experimental setup in our laboratory replicates the fully-coupled phonation process in a well-established larynx model. We aim to further extend its functionality by integrating geometry-variable simplified channels as well as an MRI-based vocal tract shape for experiments. This enables us to separately investigate the different mechanisms during phonation as influenced by a vocal tract. The project incorporates three innovative aspects. (1) The focus is set on the harmonic content and acoustic back-coupling in a fully-coupled experimental approach that represents the basic mechanisms of the phonatory process. (2) With our state-of-the-art PIV-based technique, a direct experimental determination of the acoustic source field and sound radiation is enabled. (3) A comprehensive description of the vocal fold motion and its structure interaction is generated by laser vibrometer measurements with a strong focus on its harmonic content, as well.Finally, this yields the mechanisms responsible for harmonic sound generation, which interacts with back-coupling of the vocal tract.

声音障碍具有很高的社会和经济相关性，因为它影响着我们社会中的许多人。因此，相关患者会因生活质量下降而承受更大的生理和心理压力。健康人的正常声音是复杂的流体-结构-声学相互作用的结果。声带的结构运动强烈调节通过声门的气流，声门产生声音的主要部分。因此，正常和健康的发声的特征是具有较强的谐波含量。声带下游的声道是声带运动和声流的声学反向耦合的关键因素。因此，除了产生共振峰之外，我们假设声道还会影响最终语音信号中高次谐波的产生，这对于健康的声音至关重要。该研究项目的中心目标是基于对基本机制的全耦合实验研究，增进对发声过程中音调声音生成的理解。我们实验室的实验装置在完善的喉部模型中复制了完全耦合的发声过程。我们的目标是通过集成几何变量简化通道以及基于 MRI 的声道形状进行实验来进一步扩展其功能。这使我们能够单独研究发声过程中受声道影响的不同机制。该项目包含三个创新方面。 (1) 重点放在代表发声过程基本机制的全耦合实验方法中的谐波含量和声学反向耦合。 (2) 借助我们最先进的基于 PIV 的技术，可以直接通过实验确定声源场和声辐射。 （3）通过激光振动计测量生成对声带运动及其结构相互作用的全面描述，同时重点关注其谐波内容。最后，这产生了负责谐波声音生成的机制，该机制与背景相互作用声道的耦合。