Acoustic analysis and synthesis experiments have shown that duration and intonation patterns are the two most important prosodic features responsible for the quality of synthesized speech. In this paper a set of features are proposed which will influence the duration patterns of the sequence of the sound units. These features are derived from the results of the duration analysis. Duration analysis provides a rough estimate of features, which affect the duration patterns of the sequence of the sound units. But, the prediction of durations from these features using either linear models or with a fixed rulebase is not accurate. From the analysis it is observed that there exists a gross trend in durations of syllables with respect to syllable position in the phrase, syllable position in the word, word position in the phrase, syllable identity and the context of the syllable (preceding and the following syllables). These features can be further used to predict the durations of the syllables more accurately by exploring various nonlinear models. For analying the durations of sound units, broadcast news data in Telugu is used as the speech corpus. The prediction accuracy of the duration models developed using rulebases and neural networks is evaluated using the objective measures such as percentage of syllables predicted within the specified deviation, average prediction error (µ), standard deviation (σ) and correlation coefficient (γ).
L. Mary, K. S. Rao, S. V. Gangashetty and B. Yegnanarayana, “Neural Network Models for Capturing Duration and Intonation Knowledge for Language and Speaker Identification,” International Conference on Cognitive and Neural Systems, Boston, May 2004.
S. Werner and E. Keller, “Prosodic Aspects of Speech,” Fundamentals of Speech Synthesis and Speech Recognition: Basic Concepts, State of the Art, the Future Challenges, E. Kelle Edition, John Wiley, Chichester, 1994. pp. 23-40.
A. N. Khan, S. V. Gangashetty and B. Yegnanarayana, “Syllabic Properties of Three Indian Languages: Implications for Speech Recognition and Language Identification,” International Conference on Natural Language Processing, Mysore, December 2003, pp. 125-134.
K. S. Rao and B. Yegnanarayana, “Modeling Syllable Duration in Indian Languages Using Neural Networks,” Proceedings of IEEE International Conference on Acoustics, Speech, Signal Processing, Montreal, May 2004, pp. 313-316.
W. N. Campbell, “Predicting Segmental Durations for Accommodation within a Syllable-Level Timing Frame-work,” Proceedings of European Conference on Speech Communication and Technology, Berlin, Vol. 2, Septem-ber 1993, pp. 1081-1084.
M. Vainio and T. Altosaar, “Modeling the Microprosody of Pitch and Loudness for Speech Synthesis with Neural Networks,” Proceedings of International Conference on Spoken Language Processing, Sidney, December 1998.
S. Lee, K. Hirose and N. Minematsu, “Incoporation of Prosodic Modules for Large Vocabulary Continuous Speech Recognition,” Proceedings of ISCA Workshop on Prosody in Speech Recognition and Understanding, New Jersey, 2001, pp. 97-101.
K. Ivano, T. Seki and S. Furui, “Noise Robust Speech Recognition Using F0 Contour Extract by Hough Transform,” Proceedings of International Conference on Spoken Language Processing, Denver, 2002, pp. 941-944.
L. Mary, K. S. Rao and B. Yegnanarayana, “Neural Network Classifiers for Language Identification Using Phonotactic and Prosodic Features,” Proceedings of International Conference on Intelligent Sensing and Information Processing (ICISIP), Chennai, January 2005, pp. 404-408. doi:10.1109/ICISIP.2005.1529486
S. R. R. Kumar and B. Yegnanarayana, “Significance of Durational Knowledge for Speech Synthesis in Indian Languages,” Proceedings of IEEE Region 10 Conference Convergent Technologies for the Asia-Pacific, Bombay, November 1989, pp. 486-489.