Deep neural network context embeddings for model selection in rich-context HMM synthesis

Proceedings of the Annual Conference of the International Speech Communication Association, INTERSPEECH

Volumn 2015-January, Issue , 2015, Pages 2207-2211

  Merritt, Thomas ^a   Yamagishi, Junichi ^a,b   Wu, Zhizheng ^a   Watts, Oliver ^a   King, Simon ^a  

^a UNIVERSITY OF EDINBURGH   (United Kingdom)

^b NATIONAL INSTITUTE OF INFORMATICS   (Japan)

Author keywords

Deep neural networks; Embedding; Hidden Markov model; Rich context; Speech synthesis

Indexed keywords

DECISION TREES; LINGUISTICS; MARKOV PROCESSES; SPEECH COMMUNICATION; SPEECH SYNTHESIS; TREES (MATHEMATICS); TRELLIS CODES;

CONTEXT SYNTHESIS; DEEP NEURAL NETWORKS; EMBEDDING; GUIDE SELECTION; MODEL SELECTION; PARAMETRIC SYNTHESIS; RICH CONTEXT; STATISTICAL PARAMETRIC SPEECH SYNTHESIS;

HIDDEN MARKOV MODELS;

EID: 84959122693     PISSN: 2308457X     EISSN: 19909772     Source Type: Conference Proceeding    
DOI: None     Document Type: Conference Paper

References (27)

1. S. King and V. Karaiskos, "The Blizzard Challenge 2010, " inProc. Blizzard Challenge, 2010.
2. -, "The Blizzard Challenge 2011, " in Proc. Blizzard Challenge, 2011.
3. -, "The Blizzard Challenge 2012, " in Proc. Blizzard Challenge, 2012.
4. S. King, "Measuring a decade of progress in text-to-speech, " Loquens, vol. 1, no. 1, 2014.
5. T. Toda and K. Tokuda, "A Speech Parameter Generation AlgorithmConsidering Global Variance for HMM-Based SpeechSynthesis, " IEICE Transactions on Information and Systems, vol. E90-D, no. 5, pp. 816-824, May 2007.
6. S. King, "An introduction to statistical parametric speech synthesis, "Sadhana, vol. 36, pp. 837-852, 2011.
7. K. Tokuda, T. Yoshimura, T. Masuko, T. Kobayashi, and T. Kitamura, "Speech parameter generation algorithms for HMM-basedspeech synthesis, " Proc. ICASSP, 2000.
8. H. Zen, K. Tokuda, and A. W. Black, "Statistical parametricspeech synthesis, " Speech Communication, vol. 51, no. 11, pp. 1039-1064, Nov. 2009.
9. T. Merritt, J. Latorre, and S. King, "Attributing modelling errorsinHMMsynthesis by stepping gradually from natural to modelledspeech, " in Proc. ICASSP, 2015.
10. T. Merritt and S. King, "Investigating the shortcomings of HMMsynthesis, " in Proc. 8th ISCA Speech Synthesis Workshop, 2013, pp. 165-170.
11. T. Merritt, T. Raitio, and S. King, "Investigating source and filtercontributions, and their interaction, to statistical parametricspeech synthesis, " in Proc. Interspeech, 2014, pp. 1509-1513.
12. G. E. Henter, T. Merritt, M. Shannon, C. Mayo, and S. King, "Measuring the perceptual effects of modelling assumptions inspeech synthesis using stimuli constructed from repeated naturalspeech, " in Proc. Interspeech, 2014, pp. 1504-1508.
13. Z.-J. Yan, Y. Qian, and F. K. Soong, "Rich context modeling forhigh quality HMM-based TTS, " in Proc. Interspeech, 2009, pp. 1755-1758.
14. -, "Rich-context unit selection ( RUS) approach to high qualityTTS, " in Proc. ICASSP, 2010, pp. 4798-4801.
15. S. Takamichi, T. Toda, Y. Shiga, H. Kawai, S. Sakti, and S. Nakamura, "An Evaluation of Parameter Generation Methods withRich Context Models in HMM-Based Speech Synthesis, " in Proc. Interspeech, 2012, pp. 1139-1142.
16. S. Takamichi, T. Toda, Y. Shiga, S. Sakti, G. Neubig, S. Nakamura, and S. Member, "Parameter Generation MethodsWith RichContext Models for High-Quality and Flexible Text-To-SpeechSynthesis, " Selected Topics in Signal Processing, IEEE Journalof, vol. 8, no. 2, pp. 239-250, 2014.
17. H. Liang, Y. Qian, F. K. Soong, and G. Liu, "A cross-languagestate mapping approach to bilingual (Mand arin-English) TTS, " inProc. ICASSP, 2008, pp. 4641-4644.
18. Z. Wu, C. Valentini-Botinhao, O. Watts, and S. King, "Deep neuralnetworks employing multi-task learning and stacked bottleneckfeatures for speech synthesis, " in ICASSP, 2015.
19. S. Bengio and G. Heigold, "Word Embeddings for Speech Recognition, "in Proc. Interspeech, 2014, pp. 1053-1057.
20. P. Taylor, "The target cost formulation in unit selection speechsynthesis. " in Proc. Interspeech, 2006, pp. 2038-2041.
21. J. R. Hershey and P. a. Olsen, "Approximating the Kullback-Leibler divergence between Gaussian mixture models, " in Proc. ICASSP, 2007.
22. M. Cooke, C. Mayo, and C. Valentini-Botinhao, "Hurricanenatural speech corpus, [sound], " LISTA Consortium, doi: 10. 7488/ds/140, 2013.
23. H. Kawahara, "STRAIGHT, exploitation of the other aspect ofVOCODER: Perceptually isomorphic decomposition of speechsounds, " Acoust. Sci. Technol., vol. 27, no. 6, pp. 349-353, 2006.
24. Objective measurement of active speech level, ITU RecommendationITU-T P. 56, International Telecommunication Union, Telecommunication Stand ardization Sector, Geneva, Switzerland, March 2011.
25. Method for the subjective assessment of intermediate quality levelof coding systems, ITU Recommendation ITU-R BS. 1534-1, InternationalTelecommunication Union Radiocommunication Assembly, Geneva, Switzerland, March 2003.
26. T. Merritt, J. Yamagishi, Z. Wu, O. Watts, and S. King, "Listeningtest materials for "Deep neural network context embeddings formodel selection in rich-context HMM synthesis", 2015 [dataset], "university of Edinburgh, The Centre for Speech Technology Research(CSTR), doi: 10. 7488/ds/256.
27. M. Wester, C. Valentini-Botinhao, and G. E. Henter, "Are we usingenough listeners No! an empirically-supported critique ofInterspeech 2014 TTS evaluations, " in Proc. Interspeech, 2015.