Voice Transformation Using Two-Level Dynamic Warping and Neural Networks

Voice transformation, for example, from a male speaker to a female speaker, is achieved here using a two-level dynamic warping algorithm in conjunction with an artificial neural network. An outer warping process which temporally aligns blocks of speech (dynamic time warp, DTW) invokes an inner warping process, which spectrally aligns based on magnitude spectra (dynamic frequency warp, DFW). The mapping function produced by inner dynamic frequency warp is used to move spectral information from a source speaker to a target speaker. Artifacts arising from this amplitude spectral mapping are reduced by reconstructing phase information. Information obtained by this process is used to train an artificial neural network to produce spectral warping information based on spectral input data. The performance of the speech mapping compared using Mel-Cepstral Distortion (MCD) with previous voice transformation research, and it is shown to perform better than other methods, based on their reported MCD scores.

Download Full-text

Application of Artificial Neural Network in Simulation of Supercritical Extraction of Valerenic Acid from Valeriana officinalis L.

ISRN Chemical Engineering ◽

10.5402/2012/572421 ◽

2012 ◽

Vol 2012 ◽

pp. 1-7

Author(s):

Amir Rabiee Kenaree ◽

Shohreh Fatemi

Keyword(s):

Neural Network ◽

Artificial Neural Network ◽

Supercritical Extraction ◽

The Third ◽

Marquardt Algorithm ◽

Artificial Neural ◽

Valerenic Acid ◽

Hidden Layer ◽

Dynamic Time ◽

Mlp Model

Application of artificial neural network (ANN) has been studied for simulation of the extraction process by supercritical CO2. Supercritical extraction of valerenic acid from Valeriana officianalis L. has been studied and simulated according to the significant operational parameters such as pressure, temperature, and dynamic extraction time. ANN, using multilayer perceptron (MLP) model, is employed to predict the amount of extracted VA versus the studied variables. Three tests, validation, and training data sets in three various scenarios are selected to predict the amount of extracted VA at dynamic time of extraction, working pressure, and temperature values. Levenberg-Marquardt algorithm has been employed to train the MLP network. The model in first scenario has three neurons in one hidden layer, and the models associated with the second and the third scenarios have four neurons in one hidden layer. The determination coefficients are calculated as 0.971, 0.940, and 0.964 for the first, second, and the third scenarios, respectively, demonstrating the effectiveness of the MLP model in simulating this process using any of the scenarios, and accurate prediction of extraction yield has been revealed in different working conditions of pressure, temperature, and dynamic time of extraction.

Download Full-text

The Monitoring of CNC Machine Processing Condition Based on the BP Neural Network

Advanced Materials Research ◽

10.4028/www.scientific.net/amr.542-543.95 ◽

2012 ◽

Vol 542-543 ◽

pp. 95-98 ◽

Cited By ~ 1

Author(s):

Zhong Qi Sheng ◽

Jian Yong Wang ◽

Zhi Peng Liu ◽

Liang Dong

Keyword(s):

Neural Network ◽

Artificial Neural Network ◽

Processing Condition ◽

Mapping Function ◽

Machining Process ◽

Cnc Machine ◽

Bp Artificial Neural Network ◽

Labview Software ◽

Signal Features ◽

Artificial Neural

As an example for the test of built monitoring system, the experimentation was carried out to monitor the condition of tool wear, which related with the machining process closely. In pattern recognition, using the nonlinear mapping function of neural network, a three-layer BP artificial neural network was selected to carry out the mapping between the processing status and related signal features. In this way, the monitoring of machining processing was implemented. The Matlab Script node of LabVIEW software was used to call the neural network package of Matlab, which reduced the program development cycle and increased the reliability of the program. According to the cutting experimentation of the finger milling cutter, it is verified that BP artificial neural network can effectively recognize the tool condition and accomplish the wear prediction.

Download Full-text

Surface Electrocardiogram Reconstruction From Intracardiac Electrograms Using a Dynamic Time Delay Artificial Neural Network

IEEE Transactions on Biomedical Engineering ◽

10.1109/tbme.2012.2225428 ◽

2013 ◽

Vol 60 (1) ◽

pp. 106-114 ◽

Cited By ~ 3

Author(s):

Fabienne Porée ◽

Amar Kachenoura ◽

Guy Carrault ◽

Renzo Dal Molin ◽

Philippe Mabo ◽

...

Keyword(s):

Neural Network ◽

Artificial Neural Network ◽

Time Delay ◽

Artificial Neural ◽

Dynamic Time ◽

Surface Electrocardiogram

Download Full-text

Dynamic time warping using an artificial neural network

[Proceedings 1992] IJCNN International Joint Conference on Neural Networks ◽

10.1109/ijcnn.1992.227234 ◽

2003 ◽

Cited By ~ 1

Author(s):

F.A. Unal ◽

N. Tepedelenlioglu

Keyword(s):

Neural Network ◽

Artificial Neural Network ◽

Dynamic Time Warping ◽

Time Warping ◽

Artificial Neural ◽

Dynamic Time

Download Full-text

Chaotic System Design Based on Recurrent Artificial Neural Network for the Simulation of EEG Time Series

10.4018/978-1-6684-2408-7.ch074 ◽

2022 ◽

pp. 1510-1521

Author(s):

Lei Zhang

Keyword(s):

Neural Network ◽

Time Series ◽

Artificial Neural Network ◽

Chaotic System ◽

Research Work ◽

Euler Method ◽

Open Loop ◽

Artificial Neural ◽

Artificial Neural Network Ann ◽

Dynamic Time

Electroencephalogram (EEG) signals captured from brain activities demonstrate chaotic features, and can be simulated by nonlinear dynamic time series outputs of chaotic systems. This article presents the research work of chaotic system generator design based on artificial neural network (ANN), for studying the chaotic features of human brain dynamics. The ANN training performances of Nonlinear Auto-Regressive (NAR) model are evaluated for the generation and prediction of chaotic system time series outputs, based on varying the ANN architecture and the precision of the generated training data. The NAR model is trained in open loop form with 1,000 training samples generated using Lorenz system equations and the forward Euler method. The close loop NAR model is used for the generation and prediction of Lorenz chaotic time series outputs. The training results show that better training performance can be achieved by increasing the number of feedback delays and the number of hidden neurons, at the cost of increasing the computational load.

Download Full-text

An Evaluation Model of Subgrade Stability Based on Artificial Neural Network

International Journal of Safety and Security Engineering ◽

10.18280/ijsse.100513 ◽

2020 ◽

Vol 10 (5) ◽

pp. 679-688

Author(s):

Youkun Cheng ◽

Zhenwu Shi ◽

Fajin Zu

Keyword(s):

Neural Network ◽

Artificial Neural Network ◽

Influencing Factors ◽

Evaluation Model ◽

Mapping Function ◽

Traffic Load ◽

Fuzzy Neural ◽

Artificial Neural ◽

The Law ◽

Artificial Neural Network Ann

In special areas, the highways may suffer from such diseases as deformation, cracking, subsidence, and potholes, making highway maintenance a complex and difficult task. To obtain the law-term deformation law of the subgrade and accurately evaluate the subgrade and pavement stability, this paper establishes a subgrade stability evaluation model based on artificial neural network (ANN). Firstly, the law of unstable subgrade deformation in highways of special areas was derived by analyzing the influencing factors on subgrade stability, namely, temperature field, moisture field, and traffic load. Next, the correlations between input and output characteristic quantities were extracted, and used to construct the nonredundant mapping function between influencing factors of subgrade unstable deformation and the levels of subgrade stability. Finally, a fuzzy neural network (FNN) was constructed based on Takagi-Sugeno model, realizing the evaluation of subgrade stability. The proposed model was proved effective and accurate through experiments.

Download Full-text