Influence of language parameters selection on the coarticulation of the phonemes for prosody training in TTS by neural networks

被引：0

作者：

Tucková, J ^{[1
]}

Sebesta, V ^{[1
]}

机构：

[1] Czech Tech Univ, Fac Elect Engn, CR-16635 Prague, Czech Republic

来源：

ARTIFICIAL NEURAL NETS AND GENETIC ALGORITHMS, PROCEEDINGS | 2003年

关键词：

D O I：

暂无

中图分类号：

TP18 [人工智能理论];

学科分类号：

081104 ; 0812 ; 0835 ; 1405 ;

摘要：

This contribution describes the influence of the Czech language parameters selection on the coarticulation of the phonemes for the modelling of prosody features by the artificial neural network (ANN) in a text-to-speech (TTS) synthesis. The GUHA method and neural network pruning can be used for this reason. In our work we analyzed the errors between the target and calculated values of F-0 and D from the point of view of the different context of speech units. The context of three phonemes combinations CCC, VVC, VCV, CVV, VCC, CCV, and CVC (C = consonant, V = vowel) were analyzed for the determination of a next improvement of prosody. The qualitative criteria have been found in this contribution.

引用

页码：85 / 90

页数：6

共 50 条

[31] Use of neural networks in automated selection of technological parameters of cutting tools
Universita degli Studi di Pisa, Pisa, Italy
Comput Integr Manuf Syst, 3 (137-148):
[32] Use of neural networks in automated selection of technological parameters of cutting tools
Santochi, M
Dini, G
COMPUTER INTEGRATED MANUFACTURING SYSTEMS, 1996, 9 (03): : 137 - 148
[33] LANGUAGE RECOGNITION USING DEEP NEURAL NETWORKS WITH VERY LIMITED TRAINING DATA
Ranjan, Shivesh
Yu, Chengzhu
Zhang, Chunlei
Kelly, Finnian
Hansen, John H. L.
2016 IEEE INTERNATIONAL CONFERENCE ON ACOUSTICS, SPEECH AND SIGNAL PROCESSING PROCEEDINGS, 2016, : 5830 - 5834
[34] A Weight-Selection Strategy on Training Deep Neural Networks for Imbalanced Classification
Sze-To, Antonio
Wong, Andrew K. C.
IMAGE ANALYSIS AND RECOGNITION, ICIAR 2017, 2017, 10317 : 3 - 10
[35] Neural Networks Training for Weapon Selection in First-Person Shooter Games
Petrakis, Stelios
Tefas, Anastasios
ARTIFICIAL NEURAL NETWORKS (ICANN 2010), PT III, 2010, 6354 : 417 - 422
[36] Algorithmic approaches for studies of variable influence, contribution and selection in neural networks
Andersson, FO
Åberg, M
Jacobsson, SP
CHEMOMETRICS AND INTELLIGENT LABORATORY SYSTEMS, 2000, 51 (01) : 61 - 72
[37] Selection and evaluation of input parameters of neural networks using grey superior analysis
Yin, Xiang-Gang
Yu, Weidong
TEXTILE RESEARCH JOURNAL, 2007, 77 (06) : 375 - 386
[38] Influence of Training Data on the Invertability of Neural Networks for Handwritten Digit Recognition
Adler, Antonia
Geierhos, Michaela
Hobley, Eleanor
20TH IEEE INTERNATIONAL CONFERENCE ON MACHINE LEARNING AND APPLICATIONS (ICMLA 2021), 2021, : 730 - 737
[39] Analysis of Training Parameters of Feed Forward Neural Networks for WiFi RSSI Modeling
Bogdandy, Bence
Toth, Zsolt
2019 IEEE 15TH INTERNATIONAL SCIENTIFIC CONFERENCE ON INFORMATICS (INFORMATICS 2019), 2019, : 273 - 277
[40] Training of sparse and dense deep neural networks: Fewer parameters, same performance
Chicchi, Lorenzo
Giambagli, Lorenzo
Buffoni, Lorenzo
Carletti, Timoteo
Ciavarella, Marco
Fanelli, Duccio
PHYSICAL REVIEW E, 2021, 104 (05)

← 1 2 3 4 5 →