DNN-Based Speech Synthesis: Importance of Input Features and Training Data

被引：4

作者：

Lazaridis, Alexandros ^{[1
]}

Potard, Blaise ^{[1
]}

Garner, Philip N. ^{[1
]}

机构：

[1] Idiap Res Inst, Martigny, Switzerland

来源：

SPEECH AND COMPUTER (SPECOM 2015) | 2015年 / 9319卷

关键词：

Text-to-speech synthesis; Statistical parametric synthesis; Deep neural networks; Hidden markov models;

D O I：

10.1007/978-3-319-23132-7_24

中图分类号：

TP18 [人工智能理论];

学科分类号：

081104 ; 0812 ; 0835 ; 1405 ;

摘要：

Deep neural networks (DNNs) have been recently introduced in speech synthesis. In this paper, an investigation on the importance of input features and training data on speaker dependent (SD) DNN-based speech synthesis is presented. Various aspects of the training procedure of DNNs are investigated in this work. Additionally, several training sets of different size (i.e., 13.5, 3.6 and 1.5 h of speech) are evaluated.

引用

页码：193 / 200

页数：8

共 50 条

[1] SNR-Based Features and Diverse Training Data for Robust DNN-Based Speech Enhancement
Rehr, Robert
Gerkmann, Timo
[J]. IEEE/ACM Transactions on Audio Speech and Language Processing, 2021, 29 : 1937 - 1949
[2] SNR-Based Features and Diverse Training Data for Robust DNN-Based Speech Enhancement
Rehr, Robert
Gerkmann, Timo
[J]. IEEE-ACM TRANSACTIONS ON AUDIO SPEECH AND LANGUAGE PROCESSING, 2021, 29 : 1937 - 1949
[3] On the Training of DNN-based Average Voice Model for Speech Synthesis
Yang, Shan
Wu, Zhizheng
Xie, Lei
[J]. 2016 ASIA-PACIFIC SIGNAL AND INFORMATION PROCESSING ASSOCIATION ANNUAL SUMMIT AND CONFERENCE (APSIPA), 2016,
[4] ADAPTING AND CONTROLLING DNN-BASED SPEECH SYNTHESIS USING INPUT CODES
Luong, Hieu-Thi
Takaki, Shinji
Hente, Gustav Eje
Yamagishi, Junichi
[J]. 2017 IEEE INTERNATIONAL CONFERENCE ON ACOUSTICS, SPEECH AND SIGNAL PROCESSING (ICASSP), 2017, : 4905 - 4909
[5] DNN-Based Arabic Speech Synthesis
Amrouche, Aissa
Bentrcia, Youssouf
Boubakeur, Khadidja Nesrine
Abed, Ahcene
[J]. 2022 9TH INTERNATIONAL CONFERENCE ON ELECTRICAL AND ELECTRONICS ENGINEERING (ICEEE 2022), 2022, : 378 - 382
[6] Towards minimum perceptual error training for DNN-based speech synthesis
Valentini-Botinhao, Cassia
Wu, Zhizheng
King, Simon
[J]. 16TH ANNUAL CONFERENCE OF THE INTERNATIONAL SPEECH COMMUNICATION ASSOCIATION (INTERSPEECH 2015), VOLS 1-5, 2015, : 869 - 873
[7] Unsupervised Speaker Adaptation for DNN-based Speech Synthesis using Input Codes
Takaki, Shinji
Nishimura, Yoshikazu
Yamagishi, Junichi
[J]. 2018 ASIA-PACIFIC SIGNAL AND INFORMATION PROCESSING ASSOCIATION ANNUAL SUMMIT AND CONFERENCE (APSIPA ASC), 2018, : 649 - 658
[8] AN ANALYSIS OF NOISE-AWARE FEATURES IN COMBINATION WITH THE SIZE AND DIVERSITY OF TRAINING DATA FOR DNN-BASED SPEECH ENHANCEMENT
Rehr, Robert
Gerkmann, Timo
[J]. 2019 IEEE INTERNATIONAL CONFERENCE ON ACOUSTICS, SPEECH AND SIGNAL PROCESSING (ICASSP), 2019, : 601 - 605
[9] Pre-Training of DNN-Based Speech Synthesis Based on Bidirectional Conversion between Text and Speech
Sone, Kentaro
Nakashika, Toru
[J]. IEICE TRANSACTIONS ON INFORMATION AND SYSTEMS, 2019, E102D (08) : 1546 - 1553
[10] A DNN-based emotional speech synthesis by speaker adaptation
Yang, Hongwu
Zhang, Weizhao
Zhi, Pengpeng
[J]. 2018 ASIA-PACIFIC SIGNAL AND INFORMATION PROCESSING ASSOCIATION ANNUAL SUMMIT AND CONFERENCE (APSIPA ASC), 2018, : 633 - 637

← 1 2 3 4 5 →