Design and Implementation of a Neural Network for Voiced/Unvoiced Classification for a Given System
نویسنده
چکیده
Voiced/unvoiced classification is a task from the field of acoustics to assess the vocal folds’ contribution to speech production within a given piece of sound. However, it is a difficult task, commonly approached through means of digital signal processing, which usually delivers subpar results, especially in the transition regions between the two classes. Artificial neural networks deliver results of better quality while being able to be more efficient. This paper provides best practices for the design and the implementation of an artificial neural network approach which is able to achieve better results for this particular problem . It outlines the steps to implement a multi-layer perceptron trained with back-propagation using minibatch stochastic gradient descent. The implementation was done in Octave/Matlab. Keywords—Implementation, Linear Predictive Coding, Multilayer Perceptron, Voiced/Unvoiced Classification
منابع مشابه
SNR Classification System Based on Classification of Voiced/Unvoiced Signal
This paper proposes a signal-to-noise ratio (SNR) classification system based on a classification of voiced/unvoiced signal using a time-delay neural network for noise reduction in speech that is degraded by background noises. As such, the proposed system detects voiced and unvoiced sections, then reduces the noise signal for each input frame using the time-delay neural network.
متن کاملLeast relative entropy for voiced/unvoiced speech classification
The aim of this work is to develop ajlexible and eficient approach to the classifcation of the ratio of voiced to unvoiced excitation sources in continuous speech. To achieve this aim we adopt a probabilistic neural network approach. This is accomplished by designing a multi layer perceptron classifer trained by steepest descent minimization of the Least Relative Entropy W) cost function. By us...
متن کاملGDOP Classification and Approximation by Implementation of Time Delay Neural Network Method for Low-Cost GPS Receivers
Geometric Dilution of Precision (GDOP) is a coefficient for constellations of Global Positioning System (GPS) satellites. These satellites are organized geometrically. Traditionally, GPS GDOP computation is based on the inversion matrix with complicated measurement equations. A new strategy for calculation of GPS GDOP is construction of time series problem; it employs machine learning and artif...
متن کاملOn the use of time-delay neural networks for highly accurate classification of stop consonants
Time-Delay Neural Networks (TDNN) have been shown by Waibel et al. [1] to be a good method for the classification of dynamic speech sounds such as voiced stop consonants. In this paper we discuss key issues in the design and training of a TDNN, based on a Multi-Layer Perceptron (MLP), when used for classification of the sets of voiced stop consonants (/b/, /d/, and /g/) and unvoiced stop conson...
متن کاملHigh performance text-independent speaker recognition system based on voiced/unvoiced segmentation and multiple neural nets
This paper presents a text-independent speaker recognition system based on the voiced segments of the speech signal. The proposed system uses feedforward MLP classification with only a limited amount of training and testing data and gives a comparatively high accuracy. The techniques employed are: the Rasta-PLP speech analysis for parameter estimation, a feedforward MLP for voiced/unvoiced segm...
متن کامل