With the ageing and growth of the population, some chronic diseases, such as Parkinson's disease (PD), urge the society to a health-conscious looking for better health system designs. Some recent research endeavour has been supported by solutions grounded in ubiquitous healthcare (u-Health) coupling telemedicine, context awareness and decision support capabilities. In this work, we propose a u-healthcare system to pre-diagnose PD based on the speech signal of people under voice call. The speech stream is sampled as well as processed to support the pre-diagnose using machine learning (ML). Experiments were conducted over a PD voice dataset composed of 40 individuals by using five different ML algorithms. Based on a linear Support Vector Machine (SVM) model, a false negative rate of 10% was obtained when classifying the locution of number "three".

This brief presents a fixed-point architecture based on a reconfigurable scheme for integrating several commonly used mathematical operations of speech signal processing. The proposed design can perform two transcendental mathematical operations called logarithm and powering, and three commonly used computations with similar operations named polynomial calculation, filtering, and windowing. By analyzing the adopted algorithms of the above five operations, a simplified computing unit is designed. This unit can combine six types of operations by reconfiguring the data paths, and the same multiply-add architecture can be reused for reducing the redundant usage of logic gates. The experimental results reveal that the proposed design can work at a 200-MHz clock rate, and its gate count only has 11.9k. Compared with the results of the floating-point function, the median errors of the proposed design for computing the powering and logarithmic functions are 0.57% and 0.11%, respectively. Such results indicate that this simple architecture can be effectively used in most speech processing applications.

This paper presents a low-complexity algorithm and the corresponding hardware based on the coordinated polynomial solutions for solving line spectrum pairs (LSPs). To improve the computation of LSPs, the enhanced Tschirnhaus transform (ETT) is proposed to accelerate the coordinated polynomial solution. The proposed ETT can replace fractional multiplication with addition and shift operations, so unnecessary operations are avoided. To further simplify the hardware of the ETT, three designs are presented: the preprocessing block (PPB), the iterative root-finding block (IRFB), and the closed-form solution block (CFSB). The PPB provides a design with less gate counts that can effectively transform LPCs into general-form polynomials. Such polynomials can be further decomposed into roots using the proposed IRFB based on the Birge-Vieta method. A pipeline-recursive framework is implemented in the IRFB to save calculations. To improve hardware utilization, this paper also analyzes the coefficients relationship of the ETT by introducing the data dependency graph to design the proposed functional blocks in CFSB. The experimental results show that the proposed hardware achieves a 40-fold improvement in throughput and reduces 1.16% of gate counts at the hardware synthesis level; the chip area is 1.29 mm2. The precision analysis indicates the average log spectral distance is 0.310. Moreover, the ETT in the proposed hardware only requires 29.9% of multiplication compared with the original one. Such results reveal that the proposed work is superior to the baseline work, thereby demonstrating the effectiveness of the proposed design.

This work presents a novel hardware design based on the enhanced Tschirnhaus transform (ETT) to solve the 8-order line spectral pairs (LSPs). To reduce high-complexity problems caused by fractional multiplication, the ETT is proposed to replace such operations with integer-based shift and addition operations of the original Tschirnhaus transform. Also, the data dependency graph (DDG) of the ETT is analyzed for designing hardware units and reducing computation cycles. The proposed hardware has two key blocks: the mixture computation unit (MCU) and the multiplier-free pipelined square-root unit (PSRU). The first block is designed to fast calculate multiplication and summation operations in the ETT with the use of a two-stage pipeline architecture. The second is developed to speed up square-root operations after 8-order LSPs are decomposed into two 4-order LSPs. It can also timely process the result of the first block within limited cycles. The experimental results show that compared with the Chebyshev-based research, the proposed hardware can reduce the cycle times by 98.1% and also saved about 49.7% of gate counts. In the precision evaluation, the result indicates that 95% of the computation errors are within 0.02 and proves that the proposed hardware is capable of quantizing LSPs almost as accurately as computers do. Such results reveal that the proposed work is superior to the other Chebyshev-based methods, thereby demonstrating the effectiveness of the proposed design.

In this paper, an improved music genre classification method is presented. The proposed method makes use of the wavelet package transform (WPT) and the best basis algorithm (BBA) to accurately classify and increase classification performance. It is well known that WPT can generate a wavelet decomposition that offers a richer signal analysis. In this paper, the music signal is first decomposed into approximation and detail coefficients using WPT with the best basis algorithm to minimize the Shannon entropy and maximize the representation of music signal. This paper uses the Top-Down search strategy with cost function to select the best basis. Then the proposed method could apply support vector machine (SVM) to build a music genre classifier using the mel-frequency cepstral coefficients (MFCC) and log energies extracted from the decomposition coefficients of WPT with the best basis algorithm. Finally one can perform music genre classification with the built music genre classifier. Experiments conducted on three different music datasets have shown that the proposed method can achieve higher classification accuracy than other music genre classification methods with the same experimental setup.

This paper introduces Morphlet, a new wavelet transform adapted for voice conversion purposes. The paradigm of joint time-frequency-shape analysis of discrete-time signals, possible by means of the Discrete Shape let Transform (DST), is the basis used for the construction of Morphlets. The results assure the efficacy of the proposed transform, which is able, by itself and with the help of no other tool such as a neural network, to carry out the task, totally.

This paper proposes a new music genre classification algorithm based on dynamic music frame analysis and support vector machine (SVM). The dynamic music frame analysis could cover the long-term and the short-term music genre features which can represent the time-varying behavior of music signals. The music genre features used in this paper are mel-frequency cepstral coefficient (MFCC) and log energy with dynamic frame length. The dynamic music frame analysis will be applied to train an optimized non-linear decision rule for music genre classifier via SVM. Experimental results show that the proposed new music genre classification algorithm could achieve the average classification accuracy rate of 98% for the six different music genres, including classic, dance, lullaby, Bossa, piano, and blue.

Discriminative training of Gaussian Mixture Models (GMMs) for speech or speaker recognition purposes is usually based on the gradient descent method, in which the iteration step-size, epsilon, uses to be defined experimentally. In this letter, we derive an equation to adaptively determine epsilon, by showing that the second-order Newton-Raphson iterative method to find roots of equations is equivalent to the gradient descent algorithm.

It has been shown that the 2-D cubic spline interpolation (CSI) proposed by Truong et al. is one of the best algorithms for image resampling or compression. Such a CSI algorithm together with the image coding standard, e.g., JPEG, can be used to obtain a modified image codec while still maintaining a good quality of the reconstructed image for higher compression ratios. In this paper, a fast direct computation algorithm is developed to improve the computational efficiency of the original FFT-based 2-D CSI methods. In fact, this algorithm computes the 2-D CSI directly without explicitly calculating the complex division usually needed in the FFT or Winograd discrete Fourier transform (WDFT) algorithm. In addition, this paper describes a novel way to derivate the 2-D CSI from the 1-D CSI by using the row-column method. This new fast 2-D CSI provides a regular and simple structure based upon linear correlations. Therefore, it can be implemented by the use of a modification of Kung’s pipeline structure and is naturally suitable for VLSI implementations. Experimental results show that the proposed new fast 2-D CSI algorithm can achieve almost the same CSI performance with much fewer arithmetic operations in comparison with existing efficient algorithms.

In this paper, a new algorithm based on the Tschirnhaus transforms is developed to reduce the computation complexity of the 10-order line spectrum pairs (LSP) frequencies. The first step of the proposed algorithm is to derive a quartic equation from the 1st derivative of the given 5-degree LSP polynomial. Then the extremes of the 5-degree LSP polynomial can be found by applying the Tschirnhaus transform to the above quartic equation. By the use of these extremes as the initial approximations, one can easily solve the roots of the 5-degree LSP polynomial via the Newton's method and get the accurate LSP frequencies. One of the main advantages of the proposed algorithm is the rapid root determination of a quartic equation without complex number operations and resulting in considerable computational saving. Compared to other methods, the proposed algorithm can determine the precise LSP frequencies with the lowest computational complexity.

Speaker verification is desirable widely in many speech related applications, such as automatic telephone banking and biometric security system. This paper proposes an application of line spectrum frequency (LSF), formant, and support vector machine (SVM) to develop an algorithm of text-dependent speaker verification system. First, LSF and formant are extracted from the voiced password provided by the user. Then the proposed algorithm will make use of SVM to train the speaker characteristics from these speaker features and finally generate a claimed speaker model to discriminate between the speaker and other impostors. Experiments were conducted on the real speech signals and shown the performance of the proposed algorithm yields an equal error rate (EER) of 2.12% with 8-order LSFs and formant information. In addition, both of the false acceptance rate (FAR) and the false rejection rate (FRR) are also improved remarkably.

In this paper, we propose a design of far-field single speaker localization system including noisy speech separation and enhancement with two separated channel microphones. First, we perform noisy speech separation simulations using a fast-ICA (independent component analysis) with subspace-based speech enhancement. The fast-ICA can be used to separate the two original source signals-noise and speech from their mixtures. Subspace-based speech enhancement is then utilized to reduce the residual background noise of the separated speech. Second, we calculate time-difference-of-arrival (TDOA) estimation for five directions of both right and left sides in the horizontal plane to realize speaker localization. The TDOA method estimates the time delay difference between the incoming speech signals to microphones using the average magnitude difference function (AMDF) algorithm. Finally we evaluate the localization system under various noisy environments. The experimental results reveal that the applicability of a system with noisy speech separation and enhancement, particularly for the front directions.

This paper presents a method to enhance microcalcifications and classify their borders by applying the wavelet transform. Decomposing an image and removing its low frequency sub-band the microcalcifications are enhanced. Analyzing the effects of perturbations on high frequency sub-band it’s possible to classify its borders as smooth, rugged or undefined. Results show a false positive reduction of 69.27% using a region growing algorithm.

Nowadays, fractal analysis has been successfully applied to digital speech processing, particularly for word and phoneme segmentation, which represents one of the fundamental steps in automatic speech recognition systems. The practical use of fractal analysis for this purpose should match two principles: low computational cost, to allow the use in real-time, and accuracy in the results, in order to produce a satisfactory segmentation, sending the correct data to the classifier. Aiming at meeting these two requirements, this work proposes a technique for speech segmentation based on the fractal dimension, which is obtained by using the discrete wavelet transform that avoids the use of 1/k pre-filtering. Many families of wavelets are presented and compared, and the results assure the efficacy of the proposed method.

In this paper, there are 17 types of wavelet coefficients obtained from the Matlab software and an Aurora-2 database used to evaluate which wavelet type has a better accuracy in speaker recognition. We first determine the frequency cepstral coefficient (FCC) level to form a 114-dimentional feature vector by the use of Daubechies-4 wavelet and support vector machines (SVMs) with pre-selected exponential radial basis kernel function (ERBF) and under some additional conditions. Then, average, for each wavelet, the accuracy of 42 possible combinations about the gender of speakers considered in seven kinds of experiments corresponding to two to eight speakers. The experimental results show that the best accuracy in average will be achieved by using the reverse biorthogonal-3.5 or reverse biorthogonal-3.7 wavelet. The reverse biorthogonal-3.5 wavelet is then chosen to be the proposed wavelet function for speaker recognition in terms of shorter filter length

This paper proposes an improved DCT-based image codec embedded cubic spline interpolation (CSI) with optimal quantization. It is shown in literature that CSI is superior in performance to the other interpolation functions. One of main applications of the CSI is to cooperate with standard DCT-based JPEG to obtain the modified JPEG codec and still obtain a better quality of the reconstructed image for higher compression ratios. This paper shows that the compression performance of such a modified JPEG codec can be further improved by the use of optimal quantization. The optimal quantization makes use of the rate distortion optimization algorithm to measure DCT coefficient statistic for an image generated from the CSI scheme and then construct a DCT quantization table for the given rate/distortion specification. Experimental results show that the proposed optimal quantization can achieve a better PSNR than that of the modified JPEG codec with a default quantization table

In this paper, an improved audio classification and categorization technique is presented. This technique makes use of wavelets and support vector machines (SVMs) to accurately classify and categorize audio data. When a query audio is given, wavelets are first applied to extract acoustical features such as subband power and pitch information. Then, the proposed method uses a bottom-up SVM over these acoustical features and additional parameters, such as frequency cepstral coefficients, to accomplish audio classification and categorization. A public audio database (Muscle Fish), which consists of 410 sounds in 16 classes, is used to evaluate the performances of the proposed method against other similar schemes. Experimental results show that the classification errors are reduced from 16 (8.1%) to six (3.0%), and the categorization accuracy of a given audio sound can achieve 100% in the Top 2 matches.

Although many wavelet-based pitch detection methods have been proposed in the literature, there still remains a need to investigate new wavelet-based methods for more accurate and more robust pitch determination. In this paper, an improved wavelet-based method is developed for extraction of pitch information in noisy speech. At each decomposition in the wavelet transform, an aliasing compensation algorithm is applied to approximate and detail signals, in which the distortion of aliasing due to downsampling and upsampling operations of the wavelet transform is eliminated. In addition, this paper utilizes the concept of spatial correlation function used in signal denoising to improve the performance of pitch detection in a noisy environment. It is shown in various experimental results that this new type of method has a considerable performance improvement compared with other conventional methods and wavelet-based methods.

This paper proposes a new consonant/vowel (C/V) segmentation algorithm for Mandarin speech signal. Since the Mandarin phoneme structure is a combination of a consonant (may be null) followed by a vowel, the C/V segmentation is an important part in the Mandarin speech recognition system. Based on the wavelet transform, the proposed method can directly search for the C/V segmentation point by using a product function and energy profile. The product function is generated from the appropriate wavelet and scaling coefficients of the input speech signal, and it can be applied to indicate the C/V segmentation point. With this product function and the additional verification of the energy profile, the C/V segmentation can be accurately pointed out with a low computation complexity. Experiments are provided that demonstrate the superior performance of the proposed algorithm. An overall accuracy rate of 97.2% is achieved. This algorithm is suitable for Mandarin speech recognition task.