Experiment with GMM-Based Artefact Localization in Czech Synthetic Speech

被引：7

作者：

Pribil, Jiri ^{[1
,2
]}

Pribilova, Anna ^{[3
]}

Matousek, Jindrich ^{[1
]}

机构：

[1] Univ W Bohemia, Fac Sci Appl, Dept Cybernet, Plzen 30614, Czech Republic

[2] SAS, Inst Measurement Sci, Bratislava 84104, Slovakia

[3] Slovak Univ Technol Bratislava, Inst Elect & Photon, Fac Elect Engn & Informat Technol, Bratislava 81219, Slovakia

来源：

TEXT, SPEECH, AND DIALOGUE (TSD 2015) | 2015年 / 9302卷

关键词：

Quality of synthetic speech; Text-to-speech system; GMM classification; Statistical analysis; MODELS;

D O I：

10.1007/978-3-319-24033-6_3

中图分类号：

TP18 [人工智能理论];

学科分类号：

081104 ; 0812 ; 0835 ; 1405 ;

摘要：

The paper describes an experiment with using the statistical approach based on the Gaussian mixture models (GMM) for localization of artefacts in the synthetic speech produced by the Czech text-to-speech system employing the unit selection principle. In addition, the paper analyzes influence of different number of used GMM mixtures, and the influence of setting of the frame shift during the spectral feature analysis on the resulting artefact position accuracy. Obtained results of performed experiments confirm proper function of the chosen concept and the presented artefact position localizer can be used as an alternative to the standardly applied manual localization method.

引用

页码：23 / 31

页数：9

共 50 条

[31] GMM-based procedure for multiple hypotheses testing
Zhang, Jingyi
He, Zhijian
[J]. COMMUNICATIONS IN STATISTICS-SIMULATION AND COMPUTATION, 2024, 53 (06) : 2605 - 2623
[32] Multi-frame GMM-based block quantisation of line spectral frequencies for wideband speech
So, S
Paliwal, KK
[J]. 2005 IEEE INTERNATIONAL CONFERENCE ON ACOUSTICS, SPEECH, AND SIGNAL PROCESSING, VOLS 1-5: SPEECH PROCESSING, 2005, : 121 - 124
[33] GMM-Based Molecular Serum Profiling Framework
Plechawska-Wojcik, Malgorzata
[J]. INFORMATION AND SOFTWARE TECHNOLOGIES, ICIST 2015, 2015, 538 : 57 - 70
[34] A GMM-based User Model for Knowledge Recommendation
Yang, Nian
Wang, Guoxin
Hao, Jia
Yan, Yan
Han, Hairong
[J]. 2017 3RD IEEE INTERNATIONAL CONFERENCE ON CYBERNETICS (CYBCONF), 2017, : 308 - 312
[35] Quantization for adapted GMM-based speaker verification
Tseng, Ivy H.
Verscheure, Olivier
Turaga, Deepak S.
Chaudhari, Upendra V.
[J]. 2006 IEEE International Conference on Acoustics, Speech and Signal Processing, Vols 1-13, 2006, : 653 - 656
[36] Fast adaptation of GMM-based compact models
Levy, Christophe
Linares, Georges
Bonastre, Jean-Francois
[J]. INTERSPEECH 2007: 8TH ANNUAL CONFERENCE OF THE INTERNATIONAL SPEECH COMMUNICATION ASSOCIATION, VOLS 1-4, 2007, : 2081 - 2084
[37] A GMM-Based Speaker Identification System on FPGA
Kan, Phak Len Eh
Allen, Tim
Quigley, Steven F.
[J]. RECONFIGURABLE COMPUTING: ARCHITECTURES, TOOLS AND APPLICATIONS, 2010, 5992 : 358 - 363
[38] A GMM-Based Algorithm for Classification of Radar Emitters
Gong, Xuhua
Meng, Huadong
Wang, Xiqin
[J]. ICSP: 2008 9TH INTERNATIONAL CONFERENCE ON SIGNAL PROCESSING, VOLS 1-5, PROCEEDINGS, 2008, : 2431 - 2434
[39] Testing Interval Forecasts: A GMM-Based Approach
Dumitrescu, Elena-Ivona
Hurlin, Christophe
Madkour, Jaouad
[J]. JOURNAL OF FORECASTING, 2013, 32 (02) : 97 - 110
[40] Phone-Level Prosody Modelling With GMM-Based MDN for Diverse and Controllable Speech Synthesis
Du, Chenpeng
Yu, Kai
[J]. IEEE-ACM TRANSACTIONS ON AUDIO SPEECH AND LANGUAGE PROCESSING, 2022, 30 : 190 - 201

← 1 2 3 4 5 →