Accurate Parameter Estimation for an Articulatory Speech Synthesizer with an Improved Neural Network Mapping

Neural network (NN) applications have recently been employed to extract the parameters of an articulatory speech synthesizer from a given speech signal. Results from these attempts showed that a single NN is insufficient to cover all of the possible configurations uniquely. Moreover, apart from their computational advantages, NN mapping is so far not superior to the other mapping techniques [1]. Thus there is a clear need to improve NN solution to the inverse problem. Results from our earlier experiments with an articulatory speech synthesizer have shown that the statistical characteristic of the articulatory target pattern vectors can be exploited for an improvement in the estimation performance of a Multi-Layer Perceptron (MLP) NN [2]. In this paper, the effect of the modification to the distribution characteristic of the acoustic input pattern vectors will be investigated. The theoretical background for the effect of the input distribution characteristics on neural learning has been detailed elsewhere [3]. Empirical results for a more correct estimation of articulatory speech synthesizer parameters through exploiting the behavior of the Back Propagation (BP) algorithm are focused on here.

Accurate Parameter Estimation for an Articulatory Speech Synthesizer with an Improved Neural Network Mapping

Neural network (NN) applications have recently been employed to extract the parameters of an articulatory speech synthesizer from a given speech signal. Results from these attempts showed that a single NN is insufficient to cover all of the possible configurations uniquely. Moreover, apart from their computational advantages, NN mapping is so far not superior to the other mapping techniques [1]. Thus there is a clear need to improve NN solution to the inverse problem. Results from our earlier experiments with an articulatory speech synthesizer have shown that the statistical characteristic of the articulatory target pattern vectors can be exploited for an improvement in the estimation performance of a Multi-Layer Perceptron (MLP) NN [2]. In this paper, the effect of the modification to the distribution characteristic of the acoustic input pattern vectors will be investigated. The theoretical background for the effect of the input distribution characteristics on neural learning has been detailed elsewhere [3]. Empirical results for a more correct estimation of articulatory speech synthesizer parameters through exploiting the behavior of the Back Propagation (BP) algorithm are focused on here.

Kaynak Göster

Bibtex @ { tbtkelektrik144607, journal = {Turkish Journal of Electrical Engineering and Computer Science}, issn = {1300-0632}, eissn = {1303-6203}, address = {}, publisher = {TÜBİTAK}, year = {2001}, volume = {9}, pages = {147 - 160}, doi = {}, title = {Accurate Parameter Estimation for an Articulatory Speech Synthesizer with an Improved Neural Network Mapping}, key = {cite}, author = {Altun, Halis and Yalçınöz, Tankut} }
APA Altun, H , Yalçınöz, T . (2001). Accurate Parameter Estimation for an Articulatory Speech Synthesizer with an Improved Neural Network Mapping . Turkish Journal of Electrical Engineering and Computer Science , 9 (2) , 147-160 .
MLA Altun, H , Yalçınöz, T . "Accurate Parameter Estimation for an Articulatory Speech Synthesizer with an Improved Neural Network Mapping" . Turkish Journal of Electrical Engineering and Computer Science 9 (2001 ): 147-160 <
Chicago Altun, H , Yalçınöz, T . "Accurate Parameter Estimation for an Articulatory Speech Synthesizer with an Improved Neural Network Mapping". Turkish Journal of Electrical Engineering and Computer Science 9 (2001 ): 147-160
RIS TY - JOUR T1 - Accurate Parameter Estimation for an Articulatory Speech Synthesizer with an Improved Neural Network Mapping AU - Halis Altun , Tankut Yalçınöz Y1 - 2001 PY - 2001 N1 - DO - T2 - Turkish Journal of Electrical Engineering and Computer Science JF - Journal JO - JOR SP - 147 EP - 160 VL - 9 IS - 2 SN - 1300-0632-1303-6203 M3 - UR - Y2 - 2021 ER -
EndNote %0 Turkish Journal of Electrical Engineering and Computer Science Accurate Parameter Estimation for an Articulatory Speech Synthesizer with an Improved Neural Network Mapping %A Halis Altun , Tankut Yalçınöz %T Accurate Parameter Estimation for an Articulatory Speech Synthesizer with an Improved Neural Network Mapping %D 2001 %J Turkish Journal of Electrical Engineering and Computer Science %P 1300-0632-1303-6203 %V 9 %N 2 %R %U
ISNAD Altun, Halis , Yalçınöz, Tankut . "Accurate Parameter Estimation for an Articulatory Speech Synthesizer with an Improved Neural Network Mapping". Turkish Journal of Electrical Engineering and Computer Science 9 / 2 (Şubat 2001): 147-160 .
AMA Altun H , Yalçınöz T . Accurate Parameter Estimation for an Articulatory Speech Synthesizer with an Improved Neural Network Mapping. Turkish Journal of Electrical Engineering and Computer Science. 2001; 9(2): 147-160.
Vancouver Altun H , Yalçınöz T . Accurate Parameter Estimation for an Articulatory Speech Synthesizer with an Improved Neural Network Mapping. Turkish Journal of Electrical Engineering and Computer Science. 2001; 9(2): 147-160.
IEEE H. Altun ve T. Yalçınöz , "Accurate Parameter Estimation for an Articulatory Speech Synthesizer with an Improved Neural Network Mapping", Turkish Journal of Electrical Engineering and Computer Science, c. 9, sayı. 2, ss. 147-160, Şub. 2001