数字助听器中广义旁瓣抵消器结构的汉语语音处理技术

A Chinese speech processing technology with generalized sidelobe canceller structure in hearing aids

自适应波束形成算法是数字助听器的核心算法之一。针对自适应波束形成算法中不可避免存在的语音泄漏,本文先对传统GSC结构自适应波束形成算法进行理论研究,并提出一种汉语处理技术,补偿泄漏的语音。这种汉语处理技术利用汉语语音特有的基音频率信息,调整语音幅度谱包络,提高谱包络与基频曲线形状的相似度以提高语音的可懂度。针对泄漏的语音在高频清辅音段有较大损失的特点,在频域上对清辅音进行放大,在不改变共振峰结构的情况下,提高清辅音的能量,同时降低语音间隔段GSC算法泄漏的噪声能量,提高对语音的辨别。仿真实验结果表明,这种汉语语音处理能够补偿自适应波束形成算法造成的语音泄漏,提高语音的可懂度。

Adaptive beamformer is one of the core algorithms in hearing aids. In this paper, theoretical research has been made to traditional adaptive GSC beamformer algorithm for the leak of speech existed in this algorithm. A speech processing technology was proposed to compensate the leak of speech. In this technology, amplitude spectrum envelope was modified to improve the similarity between spectrum envelope and the pitch contour with the information of pitch frequency curves so that the intelligibility of speech is improved. Considering that the speech signal has a great loss in the high frequency range, as a characteristic of the leak information, a consonant amplification method has been shown in this paper to raise the energy of voiceless consonant, reduce the noise energy and improve speech intelligibility without changing the formant structure. The simulation results show that this speech processing technology is able to compensate the leak of speech and improve the intelligibility of speech.