RESEARCH AND DEVELOPMENT OF PRACTICAL HEARING AIDS FOR TELEPHONIC SPEECH WITH A SMALL NON-LINEAR DISTORTION

小非线性失真电话语音实用助听器的研制

• 批准号: 10555130
• 负责人: IKEDA Takashi
• 金额: $ 2.43万
• 依托单位: KURUME NATIONAL COLLEGE OF TECHNOLOGY
• 依托单位国家: 日本
• 项目类别: Grant-in-Aid for Scientific Research (B).
• 财政年份: 1998
• 资助国家: 日本
• 起止时间: 1998 至 2000
• 项目状态: 已结题
• 来源: https://kaken.nii.ac.jp/en/grant/KAKENHI-PROJECT-10555130/
• 关键词: hearing aid; telephone; DSP; single resonance; 単共振分解; FPGA

To realize the single resonant analysis type hearing aid for telephonic speech by real-time processing, we tried to make a small DSP system, which is based on our system for research. Single resonant analysis (SRA) type hearing aid presents speech sounds with a small nonlinear distortion compared with other types of hearing aid, but it needs more complicated calculations than other types of hearing aids. The system for research has four DSP units to perform real-time calculation, and includes hardware for developing the algorisms for hearing aids. As the ability of one DSP improves very much these days, two DSPs at high speed (60 MFLOPS each) are used to build up compact hearing aid. And the software had fitted to new hardware. It proved to be able to reduce the volume of the hearing aid to 1/20 of research system. This size is small enough to place to the sides of a telephone. In the last part of this project, much improvement in DSP performance was reported, and one DSP hearing aid for telephonic speech can be developed for the field test of SRA type hearing aid near future. And the fitting programs also should be improved for practical tests by hearing impaired person.

为了实现单谐振分析型电话语音助听器的实时处理，我们尝试制作了一个基于本系统的小型DSP系统进行研究。与其他类型的助听器相比，单共振分析(SRA)型助听器呈现的语音具有较小的非线性失真，但它需要比其他类型的助听器更复杂的计算。该研究系统有四个DSP单元进行实时计算，并包括用于开发助听器算法的硬件。如今，随着一个DSP的能力大大提高，人们使用两个高速(每个60 MFLOPS)的DSP来构建紧凑型助听器。而且软件已经适应了新的硬件。事实证明，它能够将助听器的体积缩小到研究系统的1/20。这个尺寸足够小，可以放在电话的两侧。在本项目的最后部分，报告了在数字信号处理器性能方面的改进，并在不久的将来可以开发出一种用于电话语音的数字信号处理器助听器，用于SRA型助听器的现场测试。同时，对于听障人士的实际测试，也应改进拟合程序。