• DocumentCode
    2869957
  • Title

    Study on the distribution of syllable duration of Tibetan Lhasa dialect in different styles

  • Author

    Xiaoliang, Li ; Yincang, Li

  • Author_Institution
    Dept. of Foreign Languages, Lanzhou Commercial Coll., Lanzhou, China
  • Volume
    9
  • fYear
    2010
  • fDate
    22-24 Oct. 2010
  • Abstract
    The study of syllable duration is the basis of prosodic models and speech synthesis systems. This article does the statistical analysis on the syllable duration of Lhasa dialect from three levels: the vocabulary, rhyme style, discourses of news. The results show that: the distribution of vocabulary duration in male and female voices is concentrated; as the vocabulary duration increases, the average duration of syllable gradually becomes smaller, the rate of decline reduces; the form of love songs, verse with motto category is fixed and they have more obvious rhythm, and the difference between the average duration of syllable in different sentences is significant; the average duration of syllable in News Corpus is smallest, followed by verses, the vocabulary duration is largest; in the three levels, the average duration of syllable in female voices is larger than the average male duration. The statistical analysis on syllable duration laid a good foundation for the Tibetan language model and prosodic model.
  • Keywords
    natural language processing; speech synthesis; statistical analysis; Tibetan Lhasa dialect; Tibetan language model; love songs; news corpus; prosodic models; speech synthesis systems; statistical analysis; syllable duration distribution; verse with motto category; Computer applications; Modeling; Rhythm; Speech; Speech synthesis; Statistical analysis; Vocabulary; Tibetan Lhasa dialect; statistical analysis; syllable; syllable duration;
  • fLanguage
    English
  • Publisher
    ieee
  • Conference_Titel
    Computer Application and System Modeling (ICCASM), 2010 International Conference on
  • Conference_Location
    Taiyuan
  • Print_ISBN
    978-1-4244-7235-2
  • Electronic_ISBN
    978-1-4244-7237-6
  • Type

    conf

  • DOI
    10.1109/ICCASM.2010.5622957
  • Filename
    5622957