NeuralDPS: Neural Deterministic Plus Stochastic Model With Multiband Excitation for Noise-Controllable Waveform Generation

doi:10.1109/TASLP.2022.3140480

CORC > 自动化研究所 > 中国科学院自动化研究所 > 模式识别国家重点实验室 > 智能交互

	NeuralDPS: Neural Deterministic Plus Stochastic Model With Multiband Excitation for Noise-Controllable Waveform Generation
	Wang, Tao 1,2; Fu, Ruibo 2; Yi, Jiangyan 2; Tao, Jianhua 2; Wen, Zhengqi 2
刊名	IEEE-ACM TRANSACTIONS ON AUDIO SPEECH AND LANGUAGE PROCESSING
	2022
卷号	30 页码:865-878
关键词	Vocoders Stochastic processes Neural networks Speech processing Signal to noise ratio Acoustics Speech enhancement Vocoder speech synthesis deterministic plus stochastic multiband excitation noise control
ISSN号	2329-9290
DOI	10.1109/TASLP.2022.3140480
通讯作者	Fu, Ruibo(ruibo.fu@nlpr.ia.ac.cn) ; Yi, Jiangyan(jiangyan.yi@nlpr.ia.ac.cn) ; Tao, Jianhua(jhtao@nlpr.ia.ac.cn)
英文摘要	The traditional vocoders have the advantages of high synthesis efficiency, strong interpretability, and speech editability, while the neural vocoders have the advantage of high synthesis quality. To combine the advantages of two vocoders, inspired by the traditional deterministic plus stochastic model, this paper proposes a novel neural vocoder named NeuralDPS which can retain high speech quality and acquire high synthesis efficiency and noise controllability. Firstly, this framework contains four modules: a deterministic source module, a stochastic source module, a neural V/UV decision module and a neural filter module. The input required by the vocoder is just the spectral parameter, which avoids the error caused by estimating additional parameters, such as F0. Secondly, to solve the problem that different frequency bands may have different proportions of deterministic components and stochastic components, a multiband excitation strategy is used to generate a more accurate excitation signal and reduce the neural filter's burden. Thirdly, a method to control noise components of speech is proposed. In this way, the signal-to-noise ratio (SNR) of speech can be adjusted easily. Objective and subjective experimental results show that our proposed NeuralDPS vocoder can obtain similar performance with the WaveNet and it generates waveforms at least 280 times faster than the WaveNet vocoder. It is also 28% faster than WaveGAN's synthesis efficiency on a single CPU core. We have also verified through experiments that this method can effectively control the noise components in the predicted speech and adjust the SNR of speech.
资助项目	National Key Research and Development Plan of China[2020AAA0140002] ; National Natural Science Foundation of China (NSFC)[62101553] ; National Natural Science Foundation of China (NSFC)[61831022] ; National Natural Science Foundation of China (NSFC)[61901473] ; National Natural Science Foundation of China (NSFC)[61771472] ; Inria-CAS Joint Research Project[173211KYSB 20190049]
WOS关键词	SPEECH SYNTHESIS ; VOCODER
WOS研究方向	Acoustics ; Engineering
语种	英语
出版者	IEEE-INST ELECTRICAL ELECTRONICS ENGINEERS INC
WOS记录号	WOS:000761216400001
资助机构	National Key Research and Development Plan of China ; National Natural Science Foundation of China (NSFC) ; Inria-CAS Joint Research Project
内容类型	期刊论文
源URL	[http://ir.ia.ac.cn/handle/173211/47918]
专题	模式识别国家重点实验室_智能交互
通讯作者	Fu, Ruibo; Yi, Jiangyan; Tao, Jianhua
作者单位	1.Univ Chinese Acad Sci, Sch Artificial Intelligence, Beijing 100190, Peoples R China 2.Chinese Acad Sci, Inst Automat, Natl Lab Pattern Recognit, Beijing 100190, Peoples R China
推荐引用方式 GB/T 7714	Wang, Tao,Fu, Ruibo,Yi, Jiangyan,et al. NeuralDPS: Neural Deterministic Plus Stochastic Model With Multiband Excitation for Noise-Controllable Waveform Generation[J]. IEEE-ACM TRANSACTIONS ON AUDIO SPEECH AND LANGUAGE PROCESSING,2022,30:865-878.
APA	Wang, Tao,Fu, Ruibo,Yi, Jiangyan,Tao, Jianhua,&Wen, Zhengqi.(2022).NeuralDPS: Neural Deterministic Plus Stochastic Model With Multiband Excitation for Noise-Controllable Waveform Generation.IEEE-ACM TRANSACTIONS ON AUDIO SPEECH AND LANGUAGE PROCESSING,30,865-878.
MLA	Wang, Tao,et al."NeuralDPS: Neural Deterministic Plus Stochastic Model With Multiband Excitation for Noise-Controllable Waveform Generation".IEEE-ACM TRANSACTIONS ON AUDIO SPEECH AND LANGUAGE PROCESSING 30(2022):865-878.