Merlin 툴킷을 이용한 한국어 TTS 시스템의 심층 신경망 구조 성능 비교
Performance comparison of various deep neural network architectures using Merlin toolkit for a Korean TTS system
0
조회
0
다운로드
0
인용
0
참고문헌
키워드
초록
In this paper, we construct a Korean text-to-speech system using the Merlin toolkit which is an open source system for speech synthesis. In the text-to-speech system, the HMM-based statistical parametric speech synthesis method is widely used, but it is known that the quality of synthesized speech is degraded due to limitations of the acoustic modeling scheme that includes context factors. In this paper, we propose an acoustic modeling architecture that uses deep neural network technique, which shows excellent performance in various fields. Fully connected deep feedforward neural network (DNN), recurrent neural network (RNN), gated recurrent unit (GRU), long short-term memory (LSTM), bidirectional LSTM (BLSTM) are included in the architecture. Experimental results have shown that the performance is improved by including sequence modeling in the architecture, and the architecture with LSTM or BLSTM shows the best performance. It has been also found that inclusion of delta and delta-delta components in the acoustic feature parameters is advantageous for performance improvement.
Abstract
In this paper, we construct a Korean text-to-speech system using the Merlin toolkit which is an open source system for speech synthesis. In the text-to-speech system, the HMM-based statistical parametric speech synthesis method is widely used, but it is known that the quality of synthesized speech is degraded due to limitations of the acoustic modeling scheme that includes context factors. In this paper, we propose an acoustic modeling architecture that uses deep neural network technique, which shows excellent performance in various fields. Fully connected deep feedforward neural network (DNN), recurrent neural network (RNN), gated recurrent unit (GRU), long short-term memory (LSTM), bidirectional LSTM (BLSTM) are included in the architecture. Experimental results have shown that the performance is improved by including sequence modeling in the architecture, and the architecture with LSTM or BLSTM shows the best performance. It has been also found that inclusion of delta and delta-delta components in the acoustic feature parameters is advantageous for performance improvement.
참고문헌 (27)
- Bengio, Y., "Learning long-term dependencies with gradient descent is difficult", IEEE Transactions on Neural Networks, 5(2), pp.157~166 (1994)
- Bengio, Y., "Learning deep architectures for AI", Foundations and Trends in Machine Learning, 2(1), pp.1~127 (2009)
- "Empirical evaluation of gated recurrent neural networks on sequence modeling" (2014)[URL]
- Hochreiter, S., "Long short-term memory", Neural Computation, 9(8), pp.1735~1780 (1997)
- Hunt, A. J., "Unit selection in a concatenative speech synthesis system using a large speech database", Proceedings of the International Conference on Acoustics, pp.373~376 (1996)
- (2017)[URL]
- Kawahara, H., "Restructuring speech representations using a pitch-adaptive timefrequency smoothing and an instantaneous-frequency-based F0 extraction: Possible role of a repetitive structure in sounds", Speech Communication, 27(3-4), pp.187~207 (1999)
- Kubichek, R., "Mel-cepstral distance measure for objective speech quality assessment", Proceedings of the IEEE Pacific Rim Conference on Communications Computers and Signal Processing, pp.125~128 (1993)
- Ling, Z. H., "Deep learning for acoustic modeling in parametric speech generation : A systematic review of existing techniques & future trends", IEEE Signal Processing Magazine, 32(3), pp.35~52 (2015)
- Luo, Z., "Emotional voice conversion using deep neural networks with MCC and F0 features", Proceedings of the IEEE, 2016, 1~5
- Merritt, T., "Attributing modelling errors in HMM synthesis by stepping gradually from natural to modelled speech", Proceedings of the International Conference on Acoustics, Speech, Signal Processing, pp.4220~4224 (2015)
- Morise, M., "WORLD: A vocoder-based high-quality speech synthesis system for real-time applications", IEICE Transactions on Information and Systems, E99.D(7), pp.1877~1884 (2016)
- Najafabadi, M., "Deep learning applications and challenges in big data analytics", Journal of Big Data, 2(1), pp.1~21 (2015)
- Riedi, M., "A neural-network-based model of segmental duration for speech synthesis", Proceedings of the Eurospeech 1995, pp.599~602 (1995)
- Schuster, M., "Bidirectional recurrent neural networks", IEEE Transactions on Signal Processing, 45(11), pp.2673~2681 (1997)
- Tokuda, K., "Speech parameter generation from HMM using dynamic features", Proceedings of the 1995 International Conference on Acoustics, Speech, Signal Processing, pp.660~663 (1995)
- Weijters, T., "Speech synthesis with artificial neural networks", Proceedings of the International Conference on Neural Networks, pp.1764~1769 (1993)
- Williams, R. J., "Back-propagation: Theory, architectures and applications", pp.433~486 (1992), Lawrence Erlbaum Associates
- Wu, Z., "Merlin: An open source neural network speech synthesis system", Proceedings of the 9th ISCA Speech Synthesis Workshop, pp.202~207 (2016)
- Yoshimura, T., "Simultaneous modeling of spectrum, pitch and duration in HMM based speech synthesis", Proceedings of the Eurospeech 1999, pp.2347~2350 (1999)
- Yu, K., "Context adaptive training with factorized decision trees for HMM-based statistical parametric speech synthesis", Speech Communication, 53(6), pp.914~923 (2011)
- Zen, H., "Statistical parametric speech synthesis", Speech Communication, 51(11), pp.1039~1064 (2009)
- Zen, H., "Statistical parametric speech synthesis using deep neural networks", Proceedings of the 2013 IEEE International Conference on Acoustics, Speech, Signal Processing, pp.7962~7966 (2013)