Linear Prediction Cepstral Coefficients Research Articles

Broad Acoustic Classification of Spoken Hindi Hybrid Paired Words using Artificial Neural Networks

this paper, comparison of recognition rate on the basis of domination of vowel and consonant sound in Spoken Hindi Hybrid Paired Words (SHHPW) has been carried out, with 660 utterances as database; Linear Prediction Cepstral Coefficient (LPCC) is used as a feature extraction method and Artificial Neural Networks (ANN) as a classifier. It has been observed that Consonant dominated words provides better recognition rate as compared to vowel dominated words. Average recognition rate of 93.56% has been observed for the group with consonant dominated words on the considered data base.

Emotion recognition from speech using sub-syllabic and pitch synchronous spectral features

In this work, spectral features extracted from sub-syllabic regions and pitch synchronous analysis are proposed for speech emotion recognition. Linear prediction cepstral coefficients, mel frequency cepstral coefficients and the features extracted from high amplitude regions of spectrum are used to represent emotion specific spectral information. These features are extracted from consonant, vowel and transition regions of each syllable to study the contribution of these regions toward recognition of emotions. Consonant, vowel and the transition regions are determined using vowel onset points. Spectral features extracted from each pitch cycle, are also used to recognize emotions present in speech. The emotions used in this study are: anger, fear, happy, neutral and sad. The emotion recognition performance using sub-syllabic speech segments are compared with the results of conventional block processing approach, where entire speech signal is processed frame by frame. The proposed emotion specific features are evaluated using simulated emotion speech corpus, IITKGP-SESC (Indian Institute of Technology, KharaGPur-Simulated Emotion Speech Corpus). The emotion recognition results obtained using IITKGP-SESC are compared with the results of Berlin emotion speech corpus. Emotion recognition systems are developed using Gaussian mixture models and auto-associative neural networks. The purpose of this study is to explore sub-syllabic regions to identify the emotions embedded in a speech signal, and if possible, to avoid processing of entire speech signal for emotion recognition without serious compromise in the performance.

Emotion recognition from speech using source, system, and prosodic features

In this work, source, system, and prosodic features of speech are explored for characterizing and classifying the underlying emotions. Different speech features contribute in different ways to express the emotions, due to their complementary nature. Linear prediction residual samples chosen around glottal closure regions, and glottal pulse parameters are used to represent excitation source information. Linear prediction cepstral coefficients extracted through simple block processing and pitch synchronous analysis represent the vocal tract information. Global and local prosodic features extracted from gross statistics and temporal dynamics of the sequence of duration, pitch, and energy values represent the prosodic information. Emotion recognition models are developed using above mentioned features separately, and in combination. Simulated Telugu emotion database (IITKGP-SESC) is used to evaluate the proposed features. The emotion recognition results obtained using IITKGP-SESC are compared with the results of internationally known Berlin emotion speech database (Emo-DB). Autoassociative neural networks, Gaussian mixture models, and support vector machines are used to develop emotion recognition systems with source, system, and prosodic features, respectively. Weighted combination of evidence has been used while combining the performance of systems developed using different features. From the results, it is observed that, each of the proposed speech features has contributed toward emotion recognition. The combination of features improved the emotion recognition performance, indicating the complementary nature of the features.

DWT and LPC based feature extraction methods for isolated word recognition

In this article, new feature extraction methods, which utilize wavelet decomposition and reduced order linear predictive coding (LPC) coefficients, have been proposed for speech recognition. The coefficients have been derived from the speech frames decomposed using discrete wavelet transform. LPC coefficients derived from subband decomposition (abbreviated as WLPC) of speech frame provide better representation than modeling the frame directly. The WLPC coefficients have been further normalized in cepstrum domain to get new set of features denoted as wavelet subband cepstral mean normalized features. The proposed approaches provide effective (better recognition rate), efficient (reduced feature vector dimension), and noise robust features. The performance of these techniques have been evaluated on the TI-46 isolated word database and own created Marathi digits database in a white noise environment using the continuous density hidden Markov model. The experimental results also show the superiority of the proposed techniques over the conventional methods like linear predictive cepstral coefficients, Mel-frequency cepstral coefficients, spectral subtraction, and cepstral mean normalization in presence of additive white Gaussian noise.

Investigation of distance effect on Gaussian Mixture Models in Speaker Identification

This paper investigate the effect of distance on the Gaussian Mixture Models (GMM) for text dependent speaker identification. Three stages are used for three different distances from the microphone (1m, 2m, and 3m). The set of feature extraction used here include Mel frequency cepstral coefficient (MFCC), Bark frequency cepstral coefficient (BFCC) and linear predictive cepstral coefficient (LPCC). These features are obtained from 20 speakers (10 adults and 10 children) ;all spoke five Arabic words in 5 seconds. The set of classification includes two types GMM and multilayer perceptron neural network (MLP). Total results show that MFCC has the best performance in feature extraction, and GMM has better recognition than MLP as total recognition in GMM is 93.15% and recognition in MLP is 88.06%.The results show also that the recognition rate decreases from 93.15% to 80.82% as the distance is increased from 1m to 3m.

Classification of Speech Dysfluencies Using LPC Based Parameterization Techniques

The goal of this paper is to discuss and compare three feature extraction methods: Linear Predictive Coefficients (LPC), Linear Prediction Cepstral Coefficients (LPCC) and Weighted Linear Prediction Cepstral Coefficients (WLPCC) for recognizing the stuttered events. Speech samples from the University College London Archive of Stuttered Speech (UCLASS) were used for our analysis. The stuttered events were identified through manual segmentation and were used for feature extraction. Two simple classifiers namely, k-nearest neighbour (kNN) and Linear Discriminant Analysis (LDA) were employed for speech dysfluencies classification. Conventional validation method was used for testing the reliability of the classifier results. The study on the effect of different frame length, percentage of overlapping, value of ã in a first order pre-emphasizer and different order p were discussed. The speech dysfluencies classification accuracy was found to be improved by applying statistical normalization before feature extraction. The experimental investigation elucidated LPC, LPCC and WLPCC features can be used for identifying the stuttered events and WLPCC features slightly outperforms LPCC features and LPC features.

Application of prosody models for developing speech systems in Indian languages

In this paper we demonstrate the use of prosody models for developing speech systems in Indian languages. Duration and intonation models developed using feedforward neural networks are considered as prosody models. Labelled broadcast news data in the languages Hindi, Telugu, Tamil and Kannada is used for developing the neural network models for predicting the duration and intonation. The features representing the positional, contextual and phonological constraints are used for developing the prosody models. In this paper, the use of prosody models is illustrated using speech recognition, speech synthesis, speaker recognition and language identification applications. Autoassociative neural networks and support vector machines are used as classification models for developing the speech systems. The performance of the speech systems has shown to be improved by combining the prosodic features along with one popular spectral feature set consisting of Weighted Linear Prediction Cepstral Coefficients (WLPCCs).

AUTOMATIC SEGMENTATION OF BROADCAST AUDIO SIGNALS USING AUTO ASSOCIATIVE NEURAL NETWORKS

In this paper, we describe automatic segmentation methods for audio broadcast data. Today, digital audio applications are part of our everyday lives. Since there are more and more digital audio databases in place these days, the importance of effective management for audio databases have become prominent. Broadcast audio data is recorded from the Television which comprises of various categories of audio signals. Efficient algorithms for segmenting the audio broadcast data into predefined categories are proposed. Audio features namely Linear prediction coefficients (LPC), Linear prediction cepstral coefficients, and Mel frequency cepstral coefficients (MFCC) are extracted to characterize the audio data. Auto Associative Neural Networks are used to segment the audio data into predefined categories using the extracted features. Experimental results indicate that the proposed algorithms can produce satisfactory results.

Pattern classification models for classifying and indexing audio signals

In the age of digital information, audio data has become an important part in many modern computer applications. Audio classification and indexing has been becoming a focus in the research of audio processing and pattern recognition. In this paper, we propose effective algorithms to automatically classify audio clips into one of six classes: music, news, sports, advertisement, cartoon and movie. For these categories a number of acoustic features that include linear predictive coefficients, linear predictive cepstral coefficients and mel-frequency cepstral coefficients are extracted to characterize the audio content. The autoassociative neural network model (AANN) is used to capture the distribution of the acoustic feature vectors. Then the proposed method uses a Gaussian mixture model (GMM)-based classifier where the feature vectors from each class were used to train the GMM models for those classes. During testing, the likelihood of a test sample belonging to each model is computed and the sample is assigned to the class whose model produces the highest likelihood. Audio clip extraction, feature extraction, creation of index, and retrieval of the query clip are the major issues in automatic audio indexing and retrieval. A method for indexing the classified audio using LPCC features and k-means clustering algorithm is proposed.

Hybrid Feature and Decision Fusion Based Audio-Visual Speaker Identification in Challenging Environment

The contribution of this paper is to propose a novel approach of evaluating the performance of a noise robust audio-visual speaker identification system in challenging environment. Though the traditional HMM based audio-visual speaker identification system is very sensitive to the speech parameter variation, the proposed hybrid feature and decision fusion based audio-visual speaker identification is found to be stance and performs well for improving the robustness and naturalness of human-computerinteraction. Linear Prediction Cepstral Coefficients and Mel Frequency Cepstral Coefficients are used to extract the audio features and Active Appearance Model and Active Shape Model have been used to extract the appearance and shape based features for the facial image. Principal Component Analysis method has been used to reduce the dimensionality of large feature vector and to normalize, the vector normalization algorithm has been used. Features and decision both are fused in two different levels and finally four different classifier outputs are combined in parallel fashion to achieve the identification result. The performances of all these uni-modal and multi-modal system performance have been evaluated and compared with each other on VALID audiovisual multi-modal database, containing both vocal and visual biometric modalities.

Analysis of Infant Cry Through Weighted Linear Prediction Cepstral Coefficients and Probabilistic Neural Network

Acoustic analysis of infant cry signals has been proven to be an excellent tool in the area of automatic detection of pathological status of an infant. This paper investigates the application of parameter weighting for linear prediction cepstral coefficients (LPCCs) to provide the robust representation of infant cry signals. Three classes of infant cry signals were considered such as normal cry signals, cry signals from deaf babies and babies with asphyxia. A Probabilistic Neural Network (PNN) is suggested to classify the infant cry signals into normal and pathological cries. PNN is trained with different spread factor or smoothing parameter to obtain better classification accuracy. The experimental results demonstrate that the suggested features and classification algorithms give very promising classification accuracy of above 98% and it expounds that the suggested method can be used to help medical professionals for diagnosing pathological status of an infant from cry signals.

Speaker recognition under stressed condition

This paper presents the feature analysis and design of compensators for speaker recognition under stressed speech conditions. Any condition that causes a speaker to vary his or her speech production from normal or neutral condition is called stressed speech condition. Stressed speech is induced by emotion, high workload, sleep deprivation, frustration and environmental noise. In stressed condition, the characteristics of speech signal are different from that of normal or neutral condition. Due to changes in speech signal characteristics, performance of the speaker recognition system may degrade under stressed speech conditions. Firstly, six speech features (mel-frequency cepstral coefficients (MFCC), linear prediction (LP) coefficients, linear prediction cepstral coefficients (LPCC), reflection coefficients (RC), arc-sin reflection coefficients (ARC) and log-area ratios (LAR)), which are widely used for speaker recognition, are analyzed for evaluation of their characteristics under stressed condition. Secondly, Vector Quantization (VQ) classifier and Gaussian Mixture Model (GMM) are used to evaluate speaker recognition results with different speech features. This analysis help select the best feature set for speaker recognition under stressed condition. Finally, four VQ based novel compensation techniques are proposed and evaluated for improvement of speaker recognition under stressed condition. The compensation techniques are speaker and stressed information based compensation (SSIC), compensation by removal of stressed vectors (CRSV), cepstral mean normalization (CMN) and combination of MFCC and sinusoidal amplitude (CMSA) features. Speech data from SUSAS database corresponding to four different stressed conditions, Angry, Lombard, Question and Neutral, are used for analysis of speaker recognition under stressed condition.

Classification of audio signals using AANN and GMM

Today, digital audio applications are part of our everyday lives. Audio classification can provide powerful tools for content management. If an audio clip automatically can be classified it can be stored in an organised database, which can improve the management of audio dramatically. In this paper, we propose effective algorithms to automatically classify audio clips into one of six classes: music, news, sports, advertisement, cartoon and movie. For these categories a number of acoustic features that include linear predictive coefficients, linear predictive cepstral coefficients and mel-frequency cepstral coefficients are extracted to characterize the audio content. The autoassociative neural network model (AANN) is used to capture the distribution of the acoustic feature vectors. The AANN model captures the distribution of the acoustic features of a class, and the backpropagation learning algorithm is used to adjust the weights of the network to minimize the mean square error for each feature vector. The proposed method also compares the performance of AANN with a Gaussian mixture model (GMM) wherein the feature vectors from each class were used to train the GMM models for those classes. During testing, the likelihood of a test sample belonging to each model is computed and the sample is assigned to the class whose model produces the highest likelihood.

Multimodal person verification system using face and speech

Reliable user identification is a common requirement for almost every secure system. Biometric offer a natural and reliable solution to certain aspects of identity management by recognizing the individuals based on their inherent physical and behavior characteristics. Multimodal biometric person verification is gaining much popularity in recent years as they outperform unimodal person verification. This paper presents a person verification system using speech and face data. The verification system comprises of two classifiers whose scores are fused using sum rule after normalization. The experiments are carried out on VidTIMIT database. The experimental results show that face expert designed using Two-Dimensional Linear Discriminate Analysis and speech expert using Linear Prediction Cepstral Coefficients as feature extractor and Gaussian Mixture Model as opinion generator with 16 mixture will provide a Half Total Error Rate of 1.2%.

Automated Speaker Recognition for Home Service Robots Using Genetic Algorithm and Dempster–Shafer Fusion Technique

An automated speaker recognition system for home service robots is proposed in this paper. In an uncontrolled environment, a speech classifier should be adaptive to different users and robust to noisy environments. It is usually observed that specific features and classifiers are more appropriate to parts of the problem domain than others; therefore, we propose a self-optimizing approach in which multiple feature extraction and classification techniques are simultaneously considered. The system uses a genetic algorithm to simultaneously select features and classifier, and the results from multiple classifiers are then combined using the Dempster-Shafer theory. The set of feature extractors used here includes linear-prediction coefficients, linear-prediction cepstral coefficients, mel-frequency cepstral coefficients, and bark-frequency cepstral coefficients, and the set of classifiers includes the Gaussian mixture model, support vector machines, C4.5 decision tree, k nearest neighbors, and multilayer perceptron neural network. The WEVER-R2 home service robot is used in a typical Korean home environment to collect speech signals for evaluating the performance of the proposed system for gender and age classification. Classification results show that the performance of the proposed method consistently outperforms the individual classifiers.

System on programmable chip implementation of neural network-based isolated digit recognition system

In this article, design and implementation of an isolated digit recognition system on a NiosII soft-core processor with a CycloneII FPGA is considered. Linear predictive coefficients and Mel frequency cepstral coefficients (MFCC) are used as features and their recognition performances are compared. A multilayer perceptron (MLP) is used for classification and a self-organised feature map is employed for dimensionality reduction of features. Various MLP architectures are studied in this article and it is found that a recognition accuracy of 100% is obtained with the least computational complexity using a single-layer MLP with 10 hidden nodes for the MFCC feature. Recognition performance and convergence time for training is compared for the ‘modified’ back propagation (BP) algorithm and the BP algorithm. It is observed that the modified BP is 2.82 times faster than BP with 4% increase in recognition accuracy. A hardware accelerator is proposed for the NiosII processor for the implementation of MLP and it increases the recognition speed by a factor of 278. The technique proposed in this article is also applicable for other soft-core processors such as Microblaze and Picoblaze.

Classification of audio signals using SVM and RBFNN

In the age of digital information, audio data has become an important part in many modern computer applications. Audio classification has been becoming a focus in the research of audio processing and pattern recognition. Automatic audio classification is very useful to audio indexing, content-based audio retrieval and on-line audio distribution, but it is a challenge to extract the most common and salient themes from unstructured raw audio data. In this paper, we propose effective algorithms to automatically classify audio clips into one of six classes: music, news, sports, advertisement, cartoon and movie. For these categories a number of acoustic features that include linear predictive coefficients, linear predictive cepstral coefficients and mel-frequency cepstral coefficients are extracted to characterize the audio content. Support vector machines are applied to classify audio into their respective classes by learning from training data. Then the proposed method extends the application of neural network (RBFNN) for the classification of audio. RBFNN enables nonlinear transformation followed by linear transformation to achieve a higher dimension in the hidden space. The experiments on different genres of the various categories illustrate the results of classification are significant and effective.

Optimum HMM combined with vector Quantization for Hindi Speech word Recognition

This paper proposes an optimum speaker-independent, isolated word Hidden Markov Model (HMM) recognizer for the Hindi language. The recognition system is based on the combination of the vector quantization (VQ) technique at the acoustical level and the Markovian modeling at the recognition level. The recognizer consists of three modules – feature extraction, vector quantizer and HMM training and testing modules. The scheme proposed here firstly computes the acoustic features in terms of the Linear Predictive Cepstral LPC coefficients, Mel-Frequency Cepstral coefficients and delta MFCC along with noise and silence detection. Then, codebooks are created using VQ, and finally in the recognition phase, an optimum set of parameters are derived from different phases for getting the highest recognition score. The training and testing database consists of a set of 35 utterances of nine Indian cities/states and 35 utterances of nine digits spoken in Hindi by male and female speakers. The recognition rate was observed to be 98.61%.

LP spectra vs. Mel spectra for identification of professional mimics in Indian languages

Automatic Speaker Recognition (ASR) is an economic tool for voice biometrics because of availability of low cost and powerful processors. For an ASR system to be successful in practical environments, it must have high mimic resistance, i.e., the system should not be defeated by determined mimics which may be either identical twins or professional mimics. In this paper, we demonstrate the effectiveness of Linear Prediction (LP)-based features, viz., Linear Prediction Coefficients (LPC) and Linear Prediction Cepstral Coefficients (LPCC) over filterbank-based features such as Mel-Frequency Cepstral Coefficients (MFCC) and newly proposed Teager energy-based MFCC (T-MFCC) for the identification of professional mimics in Indian languages. Results are reported for real and fictitious experiments. On the whole, it is observed that LP-based features perform better than filterbank-based features (an average jump of 23.21% and 31.43% for fictitious experiments with professional mimic in Marathi and Hindi, respectively, whereas there is an average jump of 1.64% for real experiments with professional mimic in Hindi) and we believe that this is the first time such results on identification of professional mimics in ASR are obtained. Analysis of the results is given with the help of Mean Square Error (MSE) between training and testing utterances for mimic’s imitations for target speakers and target speakers’ normal voice. Fourier spectra and corresponding LP spectra for target speaker and its impersonations provided by professional mimic are shown to justify the results. Finally, dependence of LPC on physiological characteristics of vocal tract and its relation with respect to the problem addressed in this paper is studied.

Combination of pitch synchronous analysis and fisher criterion for speaker identification

A novel text independent speaker identification system is proposed. In the proposed system, the 12-order perceptual linear predictive cepstrum and their delta coefficients in the span of five frames are extracted from the segmented speech based on the method of pitch synchronous analysis. The Fisher ratios of the original coefficients then be calculated, and the coefficients whose Fisher ratios are bigger are selected to form the 13-dimensional feature vectors of speaker. The Gaussian mixture model is used to model the speakers. The experimental results show that the identification accuracy of the proposed system is obviously better than that of the systems based on other conventional coefficients like the linear predictive cepstral coefficients and the Mel-frequency cepstral coefficients.