DocumentCode :
2314491
Title :
Model spectrum-progression with DTW and ANN for speech synthesis
Author :
Gu, Hung-Yan ; Wu, Chang-Yi
Author_Institution :
Dept. CSIE, Nat. Taiwan Univ. of Sci. & Technol., Taipei, Taiwan
fYear :
2009
fDate :
6-9 May 2009
Firstpage :
1010
Lastpage :
1013
Abstract :
In this paper, an ANN based spectrum-progression model (SPM) is proposed. This model is intended to improve the fluency level of synthetic Mandarin speech under the situation that only a small training corpus is available. In constructing this model, first each target syllable is matched with its reference syllable by using DTW. Then, each warped path, i.e. spectrum-progression path, is time normalized to fixed dimensions, and used to train an ANN based SPM. After training, the SPM is used together with other modules such as text analysis, prosody parameter generation, and signal sample generation to synthesize Mandarin speech. Then, the synthetic speech is used to conduct perception tests. The test results show that the SPM proposed here can indeed improve the fluency level noticeably.
Keywords :
learning (artificial intelligence); natural language processing; neural nets; speech synthesis; artificial neural network; dynamic time warping; model spectrum-progression; prosody parameter generation; signal sample generation; speech synthesis; synthetic Mandarin speech; text analysis; Hidden Markov models; Natural languages; Probability density function; Scanning probe microscopy; Signal generators; Signal synthesis; Speech analysis; Speech synthesis; Testing; Text analysis;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Electrical Engineering/Electronics, Computer, Telecommunications and Information Technology, 2009. ECTI-CON 2009. 6th International Conference on
Conference_Location :
Pattaya, Chonburi
Print_ISBN :
978-1-4244-3387-2
Electronic_ISBN :
978-1-4244-3388-9
Type :
conf
DOI :
10.1109/ECTICON.2009.5137216
Filename :
5137216
Link To Document :
بازگشت