论文部分内容阅读
该文使用音高映射方法,通过选择合适的标准语音,合成出音段、音色保持不变,而只是声调变为标准语音声调的教学语音用于声调训练,减少了语音信号中声调信息之外的复杂变化带来的信息冗余与干扰。以汉语双字调的合成语音为实验材料,对日本被试进行了声调训练实验。训练结果表明:合成语音方法在声调的感知和产出的相对进步率,以及泛化产出的效果上都优于标准语音方法,远好于没有训练的对照组,大部分实验结果差异在统计上具有显著性。实验结果佐证了语音学习时存在人脑的选择性注意机制,为将合成语音方法集成到计算机辅助汉语声调教学系统,提供了实验和理论依据。
This paper uses the pitch mapping method to synthesize the sound segment by selecting the appropriate standard voice, the timbre remains the same, but the teaching voice whose tone is changed into the standard voice tone is used for the tone training and the tone information in the voice signal is reduced The complex changes brought about by the information redundancy and interference. With Chinese double-syllable synthesized speech as the experimental material, Japanese subjects were trained in tonal training. The training results show that the synthetic speech method is superior to the standard speech method in the relative rate of progress of perceived and output tonal output and the effect of generalized output, much better than the control group without training. Most of the experimental results differ in statistics On the significant. The experimental results corroborate the selective attention mechanism of human brain existing in phonetics learning, which provides experimental and theoretical basis for integrating synthesized phonetic method into computer-aided Chinese phonetic teaching system.