Shengkui Zhao

Shengkui Zhao
Alibaba Group · Department of Information Technology

PhD

About

16
Publications
1,657
Reads
How we measure 'reads'
A 'read' is counted each time someone views a publication summary (such as the title, abstract, and list of authors), clicks on a figure, or views or downloads the full-text. Learn more
127
Citations

Publications

Publications (16)
Preprint
Full-text available
Cross-lingual voice conversion (VC) is an important and challenging problem due to significant mismatches of the phonetic set and the speech prosody of different languages. In this paper, we build upon the neural text-to-speech (TTS) model, i.e., FastSpeech, and LPCNet neural vocoder to design a new cross-lingual VC framework named FastSpeech-VC. W...
Preprint
Full-text available
Deep complex U-Net structure and convolutional recurrent network (CRN) structure achieve state-of-the-art performance for monaural speech enhancement. Both deep complex U-Net and CRN are encoder and decoder structures with skip connections, which heavily rely on the representation power of the complex-valued convolutional layers. In this paper, we...
Preprint
Recent state-of-the-art neural text-to-speech (TTS) synthesis models have dramatically improved intelligibility and naturalness of generated speech from text. However, building a good bilingual or code-switched TTS for a particular voice is still a challenge. The main reason is that it is not easy to obtain a bilingual corpus from a speaker who ach...
Conference Paper
Full-text available
This paper concerns the adaptive fast finite time control of a class of nonlinear uncertain systems of which the upper bounds of the system uncertainties are unknown. By using the fast non-smooth control Lyapunov function and the method of so-called adding a power integrator merging with adaptive technique, a recursive design procedure is provided,...
Conference Paper
Full-text available
In this paper, we present a terminal sliding mode control for MIMO T-S fuzzy systems. It is shown that the concept of extreme matrices is used to determine the upper bound information of the interactions of the fuzzy subsystems first, and a terminal sliding mode controller and a set of sliding variables are then designed to enable finite time reach...
Conference Paper
Full-text available
In this paper, the conjugate gradient (CG) algorithm is modified using the RLS normal equation and new data windowing scheme. It is known that CG algorithm has fast convergence rate and numerical stability. However, the existing CG algorithms still suffer from either slow convergence or high misadjustment compared with the RLS algorithm. In this pa...
Article
In this paper, the analysis for the performance of the discrete Fourier transform LMS adaptive filter (DFT-LMS) and the discrete cosine transform LMS adaptive filter (DCT-LMS) for the Markov-2 inputs is presented. To improve the convergence property of the least mean squares (LMS) adaptive filter, the DFT-LMS and DCT-LMS preprocess the inputs with...
Conference Paper
Full-text available
This paper presents a new variable step-size transform domain LMS (NVSTDLMS) algorithm. In the algorithm, a new variable step size is timely adjusted by employing all the output errors available so far. Therefore, the memories of the previous learning information are not lost and fully used to adjust the current value of the step size. What is more...
Conference Paper
Full-text available
An adaptive fuzzy multi-surface sliding mode control is proposed for a class of nonlinear systems with mismatched uncertainties. It is shown that the Mamdani fuzzy logic-based function approximator is used to learn the system uncertainties first, and the iterative multi-surface sliding mode design is then carried out. Using the concept of terminal...
Conference Paper
Full-text available
A fast variable step-size least-mean-square algorithm (MRVSS) is proposed and analyzed in this paper. The main features of the new algorithm include the twofold. 1) It eliminates the influence of the power of the measurement noise on the steady-state misadjustment, unlike a number of variable step-size LMS algorithms previously proposed. Therefore,...
Conference Paper
Full-text available
In this paper, a sliding mode control scheme is developed for a class of complex nonlinear systems with their T-S fuzzy models. It is shown that a set of extreme fuzzy subsystems are first derived, and a constructive sliding mode control law is then developed to guarantee the stability of the closed-loop fuzzy system. Simulation results are present...
Conference Paper
Full-text available
In this paper, the modified LMS and NLMS algorithms with variable step-size are presented. It is shown that the variable step size is computed using a ratio of the sums of weighted energy of the output error with two exponential factors alpha and beta, thus the fast error convergence of the modified LMS and NLMS algorithms can then be achieved. Als...

Network

Cited By