15th International Congress of Phonetic Sciences (ICPhS-15) Barcelona, Spain August 3-9, 2003

Modified LBG Clustering Algorithms for Small Unit Inventory in Corpus-Based TTS System

Jinyoung Kim (1), Youngha Chun (1), Joohun Lee (2), Seungho Choi (3)

(1) Chonnam National University, Korea
(2) Dong-Ah College, Korea
(3) Dongshin University, Korea

Corpus-based Text-To-Speech has been actively studied for the improvement of synthesized speech heading to human-like naturalness. However, the application of TTS is very restricted due to its large database size. In this paper, to solve this problem, we propose two modified algorithms of LBG clustering algorithm (split k-means). We introduce a terminating threshold of total cost in the first modification. The number of selected inventories becomes less than target cluster number if total cost reduction is enough to end iteration process. Considering frequency information of unit instances, which is obtained during synthesizing large text corpus, makes the second modification. To consider frequency information we proposed modified cost function of MinMax commonly used in selecting centroids.
   The perceptual test results show that our algorithm achieves the successful performance with reducing most the DB size and maintaining good speech quality.

Full Paper

Bibliographic reference.  Kim, Jinyoung / Chun, Youngha / Lee, Joohun / Choi, Seungho (2003): "Modified LBG clustering algorithms for small unit inventory in corpus-based TTS system", In ICPhS-15, 2561-2564.