Acoustic Features Analysis Of Uyghur Vowel Harmony

With the development of computer technology and the information society,speech synthesis and speech recognition technology has made incrediblesynchronized research, and human-computer dialogue system related products in thepractical application of voice input system for people’s daily life, learning and work tobring the simplicity of the service.In recent years, natural voice and clarity of key technologies such as speechsynthesis research has gradually become the focus. So far, in terms of clarity andnaturalness of the Uyghur speech synthesis, linguists and abroad Uyghur prosodicaccent, co articulation, acoustic characteristics of phonemes and allophones usingtraditional methods to do some research.In this paper, Uyghur speech synthesis to improve the naturalness and claritypurposes, mainly from the actual needs of speech synthesis and speech recognitiontechnology as a starting point, use the existing " Uyghur speech acoustic parametersdatabase " corpus resources through modern experimental phonetics method onUyghur specific syllable vowel harmony acoustic characteristics were analyzed.First, select from the corpus were harmonious two-syllable word, three syllablessamples and four-syllable word harmony of the word, and its broadband vowelformant mode resonance peak, pitch, duration and intensity of sound statisticalparameters such as rhythm analyze and summarize some of the important rules andconclusions.In addition, the Uyghur two-syllable word harmony, harmony acousticcharacteristics of three-syllable word and four-syllable words were analyzedharmony, and set out the parameters of the reference range based on statistical resultsconstitute vowel harmony word.In summary, the results of these studies or waveform tiled parametric speechsynthesis system to adjust the front vowel harmony synthesis problem provides an important reference. At the same time improve the clarity and naturalness of the effectof Uyghur speech synthesis has a better effect.