Chưa xác định

Improving Naturalness of HMM-Based TTS Trained with Limited Data by Temporal Decomposition

Năm XB 2013 Tạp chí / Hội thảo IEICE Transactions on Information and Systems Volume E96.D (11) DOI / Link https://doi.org/10.1587/transinf.e96.d.2417 ↗

Tác giả

Trung-Nghia Phung

Tóm tắt

SUMMARY The most important advantage of HMM-based TTS is its highly intelligible. However, speech synthesized by HMM-based TTS is mu ffl ed and far from natural, especially under limited data conditions, which is mainly caused by its over-smoothness. Therefore, the motivation for this paper is to improve the naturalness of HMM-based TTS trained under limited data conditions while preserving its intelligibility. To achieve this motivation, a hybrid TTS between HMM-based TTS and the modi-fied restricted Temporal Decomposition (MRTD), named HTD in this paper, was proposed. Here, TD is an interpolation model of decomposing a spectral or prosodic sequence of speech into sparse event targets and dynamic event functions, and MRTD is one simplified version of TD. With a determination of event functions close to the concept of co-articulation in speech, MRTD can synthesize smooth speech and the smoothness in synthesized speech can be adjusted by manipulating event targets of MRTD. Previous studies have also found that event functions of MRTD can represent linguistic information of speech, which is important to perceive speech intelligibility, while sparse event targets can convey the non-linguistics information, which is important to perceive the naturalness of speech. Therefore, prosodic trajectories and MRTD event functions of the spectral trajectory generated by HMM-based TTS were kept unchanged to preserve the high and stable intelligibility of HMM-based TTS. Whereas MRTD event targets of the spectral trajectory generated by HMM-based TTS were rendered with an original speech database to enhance the naturalness of synthesized speech. Experimental results with small Vietnamese datasets revealed that the proposed HTD was equivalent to HMM-based TTS in terms of intelligibility but was superior to it in terms of naturalness. Further discussions show that HTD had a small footprint. Therefore, the proposed HTD showed its strong e ffi ciency under limited data conditions.

Tài liệu tham khảo

[1] [1] K. Tokuda, H. Zen, and A.W. Black, “An HMM-based speech synthesis system applied to English,” Proc. SSW, 2002.

[2] [2] J. Yamagishi, M. Tamura, T. Masuko, K. Tokuda, and T. Kobayashi, “A training method of average voice model for HMM-based speech synthesis,” IEICE Trans. Fundamentals, vol.E86-A, no.8, pp.1956-1963, Aug. 2003.

[3] [3] K. Tokuda, T. Yoshimura, T. Masuko, T. Kobayashi, and T. Kitamura, “Speech parameter generation algorithms for HMM-based speech synthesis,” Proc. ICASSP, pp.1315-1318, 2000.

[4] [4] M. Zhang, J. Tao, H. Jia, and X. Wang, “Improving HMM Based speech synthesis by reducing over-smoothing problems,” Proc. ISCSLP, pp.1-4, 2008.

[5] [5] T. Toda and K. Tokuda, “A speech parameter generation algorithm considering Global Variance for HMM-Based speech synthesis,” IEICE Trans. Inf. & Syst., vol.E90-D, no.5, pp.816-824, May 2007.

[6] [6] K. Hashimoto, S. Takaki, K. Oura, and K. Tokuda, “Overview of NIT HMM-based speech synthesis system for Blizzard Challenge 2011,” Blizzard Challenge, 2011.

[7] [7] Y. Qian, F.K. Soong, and Z. Yan, “A unified trajectory tiling approach to high quality speech rendering,” IEEE Trans. Audio Speech Language Process., vol.21, no.2, pp.280-290, 2013.

[8] [8] B.S. Atal, “Efficient coding of LPC parameters by temporal decomposition,” Proc. ICASSP, pp.81-84, 1983.

[9] [9] PC. Nguyen, T. Ochi, and M. Akagi, “Modified restricted temporal decomposition and its application to low rate speech coding,” IEICE Trans. Inf. & Syst., vol.E86-D, no.3, pp.397-405, March 2003.

[10] [10] P.N. Binh and M. Akagi, “Efficient modeling of temporal structure of speech for applications in voice transformation,” Proc. Interspeech, pp.1631-1634, 2009.

[11] [11] R.B. Chicote, J. Yamagishi, S. King, J.M. Montero, and J.M. Guarasa, “Analysis of statistical parametric and unit selection speech synthesis systems applied to emotional speech,” Speech Commun., vol.52, pp.394-404, 2010.

[12] [12] H. Kawahara, “STRAIGHT, exploration of the other aspect of VOCODER: Perceptually isomorphic decomposition of speech sounds,” Acoust. Sci & Tech., vol.27, no.6, pp.349-353, 2006.

[13] [13] H. Phe, Chinh ta Tieng Viet (Vietnamese Grammar), pp.9-15, Da Nang Publisher, 2003.

[14] [14] L.C. Mai and D.N. Duc, “Design of Vietnamese speech corpus and current status,” Proc. ISCSLP, pp.748-758, 2006.

[15] [15] TT. Vu, MC. Luong and S. Nakamura, “An HMM-based Vietnamese speech synthesis system,” Proc. O-COCOSDA, pp.116-121, 2009.

← Quay lại danh sách bài báo