论文部分内容阅读
维吾尔语属阿尔泰语系突厥语族,其共振峰频率参数是语音识别和语音合成的重要依据。首次运用实验语音学的基本理论和方法,在“维吾尔语语音声学参数数据库”的办公环境语料条件下,对维吾尔语四音节元音和谐词进行了统计分析,给出了维吾尔语元音共振峰频率参数和分布规律,并通过四音节元音和谐词实验结果,用实验数据验证了其共振峰频率分布的口耳之学规律。为参数式或波形拼接式语音合成系统中调整合成前的元音和谐问题提供了重要的参考依据。
Uyghur language belongs to the Turkic language family of Altaic languages. Its formant frequency parameter is an important basis for speech recognition and speech synthesis. Using the basic theories and methods of experimental phonetics for the first time, this paper makes a statistical analysis of the Uyghur four-syllable vowel harmony words in the office environment corpus of “Uyghur Speech Acoustics Parameter Database” Frequency parameters and distribution law of formant, and through the four-syllable vowel harmony word experimental results, with the experimental data to verify the formant frequency distribution of the law of the ear. It provides an important reference for adjusting the harmony of vowels before synthesis in parametric or waveform mosaic speech synthesis system.