Linear Prediction Cepstral Coefficients Research Articles

Speaker-specific-text based speaker verification system using spectral and phase based features

In speaker recognition tasks, one of the reasons for reduced accuracy is due to closely resembling speakers in the acoustic space. In order to increase the discriminative power of the classifier, the system must be able to use only the unique features of a given speaker with respect to his/her acoustically resembling speaker. This paper proposes a technique to reduce the confusion errors, by finding speaker-specific phonemes and formulate a text using the subset of phonemes that are unique, for speaker verification task using i-vector based approach. In this paper spectral features such as linear prediction cepstral co-efficients (LPCC), perceptual linear prediction co-efficients (PLP) and phase feature such as modified group delay are experimented to analyse the importance of speaker-specific-text in speaker verification task. Experiments have been conducted on speaker verification task using speech data of 50 speakers collected in a laboratory environment. The experiments show that the equal error rate (EER) has been decreased significantly using i-vector approach with speaker-specific-text when compared to i-vector approach with random-text using different spectral and phase based features.

Significance of Higher-Order Spectral Analysis in Infant Cry Classification

In this paper, higher-order spectral analysis is applied to infant cry signals for classification of normal infant cries from pathological infant cries. From the family of higher-order spectra, bispectrum is considered for the proposed task. Bispectrum is the Fourier transform of the third-order cumulant function. To extract features from the bispectrum, application of higher-order singular value decomposition theorem is proposed. Experimental results show the average classification accuracy of \({82.44} \pm {4.03}{ \%}\) and Matthew’s correlation coefficient (MCC) of 0.62 with proposed bispectrum features. In all of the experiments reported in this paper, support vector machine with radial basis function kernel is used as the pattern classifier. Performance of the proposed features is also compared with the state-of-the-art methods such as linear frequency cepstral coefficients, Mel frequency cepstral coefficients, perceptual linear prediction coefficients, linear prediction coefficients, linear prediction cepstral coefficients and perceptual linear prediction cepstral coefficients, and is found to be better than that given by these feature sets. The proposed bispectrum-based features are shown to be robust under signal degradation or noisy conditions at various SNR levels. Performance in the presence of noise is compared with the state-of-the-art spectral feature sets using MCC scores. In addition, effectiveness of cryunit segmentation in normal and pathological infant cry classification task is reported.

Robust speaker identification via fusion of subglottal resonances and cepstral features.

This letter investigates the use of subglottal resonances (SGRs) for noise-robust speaker identification (SID). It is motivated by the speaker specificity and stationarity of subglottal acoustics, and the development of noise-robust SGR estimation algorithms which are reliable at low signal-to-noise ratios for large datasets. A two-stage framework is proposed which combines the SGRs with different cepstral features. The cepstral features are used in the first stage to reduce the number of target speakers for a test utterance, and then SGRs are used as complementary second-stage features to conduct identification. Experiments with the TIMIT and NIST 2008 databases show that SGRs, when used in conjunction with power-normalized cepstral coefficients and linear prediction cepstral coefficients, can improve the performance significantly (2%-6% absolute accuracy improvement) across all noise conditions in mismatched situations.

Wavelet based feature combination for recognition of emotions

Abstract In this paper, authors tried to develop reduced combinational features for emotional speech recognition. The spectral/cepstral features like wavelet coefficient, LPCC (linear prediction cepstral coefficient) and MFCC (mel-frequency cepstral coefficient) are used as baseline features. The frequency variation with respect to time has been evaluated properly using wavelet coefficients. Again, MFCC and LPCC features are derived from wavelets, so that the exact information of the emotion has been fetched and are used as features. The features are reduced with vector quantization method and used in radial basis function network (RBFNN) classifier. The feature sets are also combined and tested in the classifier. This piece of work deals with five emotions as angry, fear, happy, disgust and neutral. These are tested for Berlin (EMO-DB) and Surrey Audio-Visual Expressed Emotion (SAVEE) database. The proposed frequency based decomposition and combination choice of features show excellent result and it is exhibited in result section.

Intra- and Inter-database Study for Arabic, English, and German Databases: Do Conventional Speech Features Detect Voice Pathology?

A large population around the world has voice complications. Various approaches for subjective and objective evaluations have been suggested in the literature. The subjective approach strongly depends on the experience and area of expertise of a clinician, and human error cannot be neglected. On the other hand, the objective or automatic approach is noninvasive. Automatic developed systems can provide complementary information that may be helpful for a clinician in the early screening of a voice disorder. At the same time, automatic systems can be deployed in remote areas where a general practitioner can use them and may refer the patient to a specialist to avoid complications that may be life threatening. Many automatic systems for disorder detection have been developed by applying different types of conventional speech features such as the linear prediction coefficients, linear prediction cepstral coefficients, and Mel-frequency cepstral coefficients (MFCCs). This study aims to ascertain whether conventional speech features detect voice pathology reliably, and whether they can be correlated with voice quality. To investigate this, an automatic detection system based on MFCC was developed, and three different voice disorder databases were used in this study. The experimental results suggest that the accuracy of the MFCC-based system varies from database to database. The detection rate for the intra-database ranges from 72% to 95%, and that for the inter-database is from 47% to 82%. The results conclude that conventional speech features are not correlated with voice, and hence are not reliable in pathology detection.

Excavation Equipment Recognition Based on Novel Acoustic Statistical Features.

Excavation equipment recognition attracts increasing attentions in recent years due to its significance in underground pipeline network protection and civil construction management. In this paper, a novel classification algorithm based on acoustics processing is proposed for four representative excavation equipments. New acoustic statistical features, namely, the short frame energy ratio, concentration of spectrum amplitude ratio, truncated energy range, and interval of pulse are first developed to characterize acoustic signals. Then, probability density distributions of these acoustic features are analyzed and a novel classifier is presented. Experiments on real recorded acoustics of the four excavation devices are conducted to demonstrate the effectiveness of the proposed algorithm. Comparisons with two popular machine learning methods, support vector machine and extreme learning machine, combined with the popular linear prediction cepstral coefficients are provided to show the generalization capability of our method. A real surveillance system using our algorithm is developed and installed in a metro construction site for real-time recognition performance validation.

Performance of speaker identification using CSM and TM

The main objective of this paper is to develop the system of speaker identification. Speaker identification is a technology that allows a computer to automatically identify the person who is speaking, based on the information received from speech signal. One of the most difficult problems in speaker recognition is dealing with noises. The performance of speaker recognition using close speaking microphone (CSM) is affected in background noises. To overcome this problem throat microphone (TM) which has a transducer held at the throat resulting in a clean signal and unaffected by background noises is used. Acoustic features namely linear prediction coefficients, linear prediction cepstral coefficients, Mel frequency cepstral coefficients and relative spectral transform-perceptual linear prediction are extracted. These features are classified using RBFNN and AANN and their performance is analyzed. A new method was proposed for identification of speakers in clean and noisy using combined CSM and TM. The identification performance of the combined system is increased than individual system due to complementary nature of CSM and TM.

Stop Consonant-Short Vowel (SCSV) Classification for Tamil Speech Utterances

Tamil Language is one of the ancient Dravidian languages spoken in south India. Most of the Indian languages are syllabic in nature and syllables are in the form of Consonant-Vowel (CV) units. In Tamil language, CV pattern occurs in the beginning, middle and end of a word. In this work, CV units formed with Stop Consonant – Short Vowel (SCSV) were considered for classification task. The work carried out in three stages, Vowel Onset Point (VOP) detection, CV segmentation and classification. VOP is an event at which the consonant part ends and vowel part begins. VOPs are identified using linear Prediction residuals which provide significant characteristics of the excitation source. To segment the CV units, fixed length spectral frames before and after VOPs are considered. Both production based features - Linear Predictive Cepstral Coefficients (LPCC) and perception based features - Mel Frequency Cepstral Coefficients (MFCC) are extracted and given as input to the classifiers designed with multilayer perceptron and support vector machine. A speech corpus of 200 Tamil words uttered by 15 native speakers was used, which covers all SCSV units formed with Tamil stop consonants (/k/,/ch/,/d/,/t/,/p/) and short vowels (/a/,/i/, /u/, /e/, /o/). The classifiers are trained and tested for its performance using various measures. The results indicate that the model built with MFCC using support vector machine RBF kernel outperforms.

An autonomous system of detecting and attracting leafhopper males using species- and sex-specific substrate borne vibrational signals

In leafhoppers that are among the most important vectors of plant diseases, mate recognition and location are mediated exclusively by species- and sex-specific vibrational signals exchanged in precisely coordinated duets. These pests are currently managed primarily by insecticide treatments, however, current legislation and consumers’ concerns and demands require that the risks and impacts of pesticides be reduced. We present a proof-of-concept low-cost autonomous digital processing system (AS), capable of recognizing the male calls of the leafhopper Aphrodes bicincta “Dragonja” and generating female replies. Such a device could be used as a vibrational trap. We chose this species since its duet structure is complex, with the female replies having to appear in short (47–175ms) intervals between continuously repeated elements in the male call in order to trigger male searching behaviour. The AS male call recognition algorithm is based on linear prediction cepstral coefficient (LPCC) feature vectors and a multilayer perceptron classifier (MLP). To prevent the noise-based feature vectors from feeding into the classifier, a bandwidth-limited linear prediction call activity detector based on spectrum peak tracking was designed. We tested the efficiency of the AS in behavioural experiments with live males. The MLP classification method successfully classified vibrational calls of male A. bicincta “Dragonja” from background noise. The fast real time identification enabled a synchronized playback of female vibrational reply with latencies as short as 130ms. This mimicking of a duetting female by autonomous system also attracted the males to the source of the female reply. The AS is also a useful tool to enable further studies of vibrational duets that are needed to develop effective alternative control strategies.

RECOGNITION OF TAMIL SYLLABLES USING VOWEL ONSET POINTS WITH PRODUCTION, PERCEPTION BASED FEATURES

Tamil Language is one of the ancient Dravidian languages spoken in south India. Most of the Indian languages are syllabic in nature and syllables are in the form of Consonant-Vowel (CV) units. In Tamil language, CV pattern occurs in the beginning, middle and end of a word. In this work, CV Units formed with Stop Consonant - Short Vowel (SCSV) were considered for classification task. The work carried out in three stages, Vowel Onset Point (VOP) detection, CV segmentation and classification. VOP is an event at which the consonant part ends and vowel part begins. VOPs are identified using linear prediction residuals which provide significant characteristics of the excitation source. To segment the CV units, fixed length spectral frames before and after VOPs are considered. In this work, production based features, Linear Predictive Cepstral Coefficients (LPCC) and perception based features, Perceptual Linear Predictive Cepstral Coefficients (PLP) and Mel Frequency Cepstral Coefficients (MFCC) are extracted which are used to build the SCSV classifier using multilayer perceptron and support vector machine. A speech corpus of 200 Tamil words uttered by 15 native speakers was used, which covers all SCSV units formed with Tamil stop consonants (/k/, /ch/, /d/, /t/, /p/) and short vowels (/a/, /i/, /u/, /e/, /o/). The classifiers are trained and tested for its performance using predictive accuracy measure. The results indicate that perception based features, MFCC and PLP provides better results than production based features, LPCC and the model built using support vector machine outperforms.

Physiologically Motivated Feature Extraction for Robust Automatic Speech Recognition

In this paper, a new method is presented to extract robust speech features in the presence of the external noise. The proposed method based on two-dimensional Gabor filters takes in account the spectro-temporal modulation frequencies and also limits the redundancy on the feature level. The performance of the proposed feature extraction method was evaluated on isolated speech words which are extracted from TIMIT corpus and corrupted by background noise. The evaluation results demonstrate that the proposed feature extraction method outperforms the classic methods such as Perceptual Linear Prediction, Linear Predictive Coding, Linear Prediction Cepstral coefficients and Mel Frequency Cepstral Coefficients.

Automatic continuous speech recogniser for Dravidian languages using the auto associative neural network

In recent times with the extensive improvement of computers, numerous methods of data interchange between man and computer are revealed. It aims to provide an efficient way for human to communicate with computers exclusively for people with disabilities who face diversity of obstacles while using computers. This paper predominantly focuses on developing an efficient speech recognition system for Dravidian languages such as Tamil, Malayalam, Telugu and Kannada. The proposed CSR system comprises of four steps namely pre-processing, feature extraction, automatic continuous speech segmentation and classification. The most powerful and widely used short term energy and zero crossing rate is used for continuous speech segmentation and Mel frequency cepstral coefficients MFCC, linear predictive cepstral coefficients LPCC and shifted delta cepstrum SDC feature extractions are used for recognition system. Experiments are carried out with real time Dravidian languages speech signal. It is observed from the results that the proposed system gives significant results in AANN classifier with MFCC feature when compared with LPCC and SDC features.

LPC AND ITS DERIVATIVES FOR STUTTERED SPEECH RECOGNITION

Stuttering or stammering is disruptions in the normal flow of speech by dysfluencies, which can be repetitions or prolongations of phoneme or syllable. Stuttering cannot be permanently cured, though it may go into remission or stutterers can learn to shape their speech into fluent speech with an appropriate speech pathology treatment. Linear Prediction Coefficient (LPC), Linear Prediction Cepstral Coefficient (LPCC) and Line Spectral Frequency (LSF) were used for the feature extraction, while Multilayer Perceptron (MLP) was used as the classifier. The samples used were obtained from UCLASS (University College London Archive of Stuttered Speech) release 1. The LPCC-MLP system had the highest overall sensitivity, precision and the lowest overall misclassification rate. LPCC-MLP system had challenges with F3, the sensitivity of the system to F3 was negligible, similarly, the precision was moderate and the misclassification rate was negligible, but above 10%.

Digit Recognition based on Euclidean and DTW

paper describes the implementation of two isolated digit recognition techniques and is a comparison between the algorithms implemented. Any digit recognition comprises of mainly two stages feature extraction and similarity evaluation. Here, two feature extraction techniques, namely linear predictive cepstral coefficients (LPCC) and mel frequency cepstral coefficients (MFCC) are implemented and the similarity evaluation is done using Euclidean distance and Dynamic Time Warping (DTW). In DTW both single and averaged template matching is done. The results obtained for these algorithms are perused, compared and conclusions are drawn.

Comparative study of singing voice detection methods

Detecting Singing segments in a segment of a soundtrack is an important and useful technique in musical signal processing and retrieval. In this paper, we study the accuracy of detecting singing segments using the HMM (Hidden Markov Model) classifier with various features, including MFCC (Mel Frequency Cepstral Coefficients), LPCC (Linear Predictive Cepstral Coefficients), and LPC (Linear Prediction Coefficients). Simulation results show that detecting singing segments in a soundtrack is more difficult than detecting them among pure-instrument segments. In addition, combining MFCC and LPCC yield higher accuracy. The bootstrapping technique has only limited accuracy improvement to detect all singing segments in a soundtrack. To be complete, we also conduct an experiment to show that the time to perform music identification can be reduced by more than 40 % if we incorporate the singing-voice detection mechanism into the identification process.

A Study of Acoustic Features for Emotional Speaker Recognition in I-Vector Representation

Recently recognition of emotions became very important in the field of speech and/or speaker recognition. This paper is dedicated to experimental investigation of best acoustic features obtained for purpose of gender-dependent speaker recognition from emotional speech. Four feature sets LPC (Linear Prediction Coefficients), LPCC (Linear Prediction Cepstral Coefficients), MFCC (Melfrequency Cepstral Coefficients) and PLP (Perceptual linear prediction) coefficients were compared in an experimental setup of speaker recognition system, based on i-vector representation. For evaluation of the system emotional speech recordings from newly created Slovak emotional database and Mahalanobis distance metric as scoring method were used. The results of the experiment showed the MFCC representation as the best fitted for speaker verification from Slovak emotional speech with recognition rate higher than 80%.

Memory‐efficient buffering method and enhanced reference template for embedded automatic speech recognition system

This work realises a memory-efficient embedded automatic speech recognition (ASR) system on a resource-constrained platform. A buffering method called ultra-low queue-accumulator buffering is presented to efficiently use the constrained memory to extract the linear prediction cepstral coefficient (LPCC) feature in the embedded ASR system. The optimal order of the LPCC is evaluated to balance the recognition accuracy and the computational cost. In the decoding part, the proposed enhanced cross-words reference templates (CWRTs) method is incorporated into the template matching method to reach the speaker-independent characteristic of ASR tasks without the large memory burden of the conventional CWRTs method. The proposed techniques are implemented on a 16-bit microprocessor GPCE063A platform with a 49.152 MHz clock, using a sampling rate of 8 kHz. Experimental results demonstrate that recognition accuracy reaches 95.22% in a 30-sentence speaker-independent embedded ASR task, using only 0.75 kB RAM.

Noise Robust Speech Parameterization using Relative Spectra and Auditory Filterbank

In the present study, a new feature extraction method based on relative spectra and gammachirp auditory filterbank is proposed for robust noisy speech recognition. The relative spectra filtering are applied to the log of the output of the gammachirp filterbank which incorporates the properties of the cochlear filter in order to remove uncorrelated additive noise components. The performances of this method have been evaluated on the isolated speech word corrupted by real-world noisy environments using the continuous Gausian-Mixture density Hidden Markov Model. The evaluation of the experimental results shows that the proposed method achieves best recognition rates compared to the conventional techniques like Perceptual Linear Prediction (PLP), Linear Predictive Cepstral Coefficients (LPCC) and Mel-Frequency Cepstral Coefficients (MFCC).

Spectral and cepstral analyses for Parkinson's disease detection in Spanish vowels and words

AbstractAbout 1% of people older than 65 years suffer from Parkinson's disease (PD) and 90% of them develop several speech impairments, affecting phonation, articulation, prosody and fluency. Computer‐aided tools for the automatic evaluation of speech can provide useful information to the medical experts to perform a more accurate and objective diagnosis and monitoring of PD patients and can help also to evaluate the correctness and progress of their therapy. Although there are several studies that consider spectral and cepstral information to perform automatic classification of speech of people with PD, so far it is not known which is the most discriminative, spectral or cepstral analysis. In this paper, the discriminant capability of six sets of spectral and cepstral coefficients is evaluated, considering speech recordings of the five Spanish vowels and a total of 24 isolated words. According to the results, linear predictive cepstral coefficients are the most robust and exhibit values of the area under the receiver operating characteristic curve above 0.85 in 6 of the 24 words.

Particle swarm optimization based feature enhancement and feature selection for improved emotion recognition in speech and glottal signals.

In the recent years, many research works have been published using speech related features for speech emotion recognition, however, recent studies show that there is a strong correlation between emotional states and glottal features. In this work, Mel-frequency cepstralcoefficients (MFCCs), linear predictive cepstral coefficients (LPCCs), perceptual linear predictive (PLP) features, gammatone filter outputs, timbral texture features, stationary wavelet transform based timbral texture features and relative wavelet packet energy and entropy features were extracted from the emotional speech (ES) signals and its glottal waveforms(GW). Particle swarm optimization based clustering (PSOC) and wrapper based particle swarm optimization (WPSO) were proposed to enhance the discerning ability of the features and to select the discriminating features respectively. Three different emotional speech databases were utilized to gauge the proposed method. Extreme learning machine (ELM) was employed to classify the different types of emotions. Different experiments were conducted and the results show that the proposed method significantly improves the speech emotion recognition performance compared to previous works published in the literature.