Standard Mel-frequency Cepstral Coefficients Research Articles

Significance of relative phase features for shouted and normal speech classification

Shouted and normal speech classification plays an important role in many speech-related applications. The existing works are often based on magnitude-based features and ignore phase-based features, which are directly related to magnitude information. In this paper, the importance of phase-based features is explored for the detection of shouted speech. The novel contributions of this work are as follows. (1) Three phase-based features, namely, relative phase (RP), linear prediction analysis estimated speech-based RP (LPAES-RP) and linear prediction residual-based RP (LPR-RP) features, are explored for shouted and normal speech classification. (2) We propose a new RP feature, called the glottal source-based RP (GRP) feature. The main idea of the proposed GRP feature is to exploit the difference between RP and LPAES-RP features to detect shouted speech. (3) A score combination of phase- and magnitude-based features is also employed to further improve the classification performance. The proposed feature and combination are evaluated using the shouted normal electroglottograph speech (SNE-Speech) corpus. The experimental findings show that the RP, LPAES-RP, and LPR-RP features provide promising results for the detection of shouted speech. We also find that the proposed GRP feature can provide better results than those of the standard mel-frequency cepstral coefficient (MFCC) feature. Moreover, compared to using individual features, the score combination of the MFCC and RP/LPAES-RP/LPR-RP/GRP features yields an improved detection performance. Performance analysis under noisy environments shows that the score combination of the MFCC and the RP/LPAES-RP/LPR-RP features gives more robust classification. These outcomes show the importance of RP features in distinguishing shouted speech from normal speech.

Application and Improvement of MFCC in Gesture Recognition with Surface Electromyography

As a physiological signal reflecting the state of muscle activation, surface electromyography (sEMG) plays a vital role in the assessment of neuromuscular health, human–computer interaction, and gait analysis. Inspired by the audio signal analysis outcome that features extracted with Mel Frequency Cepstral Coefficient (MFCC) empower better representation, this paper proposes a comparative study of a gesture recognition method by using and improving with the MFCC features of sEMG. Comparing and combining with the conventional time-domain and frequency-domain features, different learning-based techniques are deployed to evaluate the performance of the proposed approach on the NinaPro datasets. The proposed approach was evaluated on the NinaPro-DB1 and NinaPro-DB2 datasets, achieving the improvements of 3.42% and 3.67%, respectively, in terms of the highest accuracy using the standard MFCC method. Correspondingly, when combined with the improved MFCC, the accuracy was further increased, reaching the maximum values of 89.82% and 87.82%, respectively, on the two datasets. The impact on the performance reveals the effectiveness of MFCC, and the results show that the proposed method has the potential to realize high-precision gesture recognition.

Cochlea-inspired speech recognition interface.

Automatic speech recognition (ASR) technology provides a natural interface for human-machine interaction. Typical ASR systems can achieve high performance in quiet environments but, unlike humans, perform poorly in real-world situations. To better simulate the human auditory periphery and improve the performance in realistic noisy scenarios, we propose two models of speech recognition front-ends based on a biophysical cochlear model. The first front-end is based on the method of signal reconstruction from a basilar membrane response. When applied to noisy speech, this method results in improved signal quality. This method can be used as a preprocessing step in a standard ASR system and can also be used as a noise reduction technique for other applications. The second front-end we propose is based on the construction of speech recognition coefficients directly from a basilar membrane response. Experimental results using a continuous-density hidden Markov model (HMM) recognizer demonstrate significant improvement in performance compared to standard Mel-frequency cepstral coefficients (MFCC) in various types of noisy conditions. Graphical Abstract Speech recognition model based on cochlear front-end.

Robust Cochlear-Model-Based Speech Recognition

Accurate speech recognition can provide a natural interface for human–computer interaction. Recognition rates of the modern speech recognition systems are highly dependent on background noise levels and a choice of acoustic feature extraction method can have a significant impact on system performance. This paper presents a robust speech recognition system based on a front-end motivated by human cochlear processing of audio signals. In the proposed front-end, cochlear behavior is first emulated by the filtering operations of the gammatone filterbank and subsequently by the Inner Hair cell (IHC) processing stage. Experimental results using a continuous density Hidden Markov Model (HMM) recognizer with the proposed Gammatone Hair Cell (GHC) coefficients are lower for clean speech conditions, but demonstrate significant improvement in performance in noisy conditions compared to standard Mel-Frequency Cepstral Coefficients (MFCC) baseline.

Block Energy Based Visual Features Using Histogram Of Oriented Gradient For Bimodal Hindi Speech Recognition

In this paper, new set of visual speech feature based on Histogram of Oriented Gradient (HOG) is proposed to improve the robustness of bimodal Hindi speech recognition. For extracting the visual features, energy per block using HOG is calculated by finding the gradient magnitude of the pixel in the cell for both x and y direction form the region of interest (ROI). The advantage of proposed scheme is that it has reduced the dimensionality of visual features vectors which can retain the full information of the lip region. For comparative study, four sets of visual feature; Set A (Two-Dimensional Discrete Cosine Transform feature (2D-DCT)), Set B (Two-Dimensional Discrete Wavelet Transform followed by DCT (2D-DWT-DCT)), Set C (Static-HOG) and Set D (Dynamic-HOG) are extracted from AMUAV corpus. Standard Mel Frequency Cepstral coefficients (MFCC) followed by static and dynamic (MFCC) features were used as baseline features. The maximum improvement in WRA (%) of 12.73% is reported over baseline features using proposed features sets.

Interface of an Automatic Recognition System for Dysarthric Speech

This paper addresses the realization of a Human/Machine (H/M) interface including a system for automatic recognition of the Continuous Pathological Speech (ARSCPS) and several communication tools in order to help frail people with speech problems (Dysarthric speech) to access services providing by new technologies of information and communication (TIC) while making it easier for the doctors to achieve a first diagnosis on the patient’s disease. In addition, an ARSCPS has been improved and developed for normal and pathology voice while establishing a link with our graphic interface which is based on the box tools Hidden Markov Model Toolkit (HTK), in addition to the Hidden Models of Markov (HMM). In our work we used different techniques of feature extraction for the speech recognition system in order to improve the dysarthric speech intelligibility while developing an ARSCPS which can perform well for pathological and normal speakers. These techniques are based on the coefficients of ETSI standard Mel Frequency Cepstral Coefficient Front End (ETSI MFCC FE V2.0); Perceptual Linear Prediction coefficients (PLP); Mel Frequency Cepstral Coefficients (MFCC) and the recently proposed Power Normalized Cepstral Coefficients (PNCC) have been used as a basis for comparison. In this context we used the Nemours database which contains 11 speakers that represents dysarthric speech and 11 speakers that represents normal speech.

Data selection for i-vector based automatic speaker verification anti-spoofing

State-of-the-art i-vector based automatic speaker verification (ASV) systems lead to considerably high performance and thus voice becomes one of the most important biometric modality for person authentication. However, similar to other biometrics, ASV systems are highly vulnerable to spoofing attacks. Therefore, developing countermeasures for detecting spoofing attacks plays an important role against the concerns regarding the reliability of ASV systems. Recent studies have shown that simple Gaussian mixture model (GMM) classifier outperforms i-vector countermeasures. In this study, we focus on improving the spoofing detection performance of i-vector system using cosine and probabilistic linear discriminant analysis (PLDA) scoring. Experimental results conducted on ASVspoof 2015 database reveals that the data used to train the two key elements of i-vector system, universal background model (UBM) and the i-vector extractor (T-matrix), play an important role on spoofing detection performance. In this paper, we study the effect of using different type of data (genuine/human or spoofed) to train these two elements and their performance on spoofing detection. In particular, extracting i-vectors using UBM trained with genuine (human) speech utterances and T-matrix trained from both genuine and spoofed utterances leads to 50% performance improvement on spoofing detection. With the proposed scheme, unlike the previous results, i-vector countermeasure outperforms GMM classifier. Finally, experimental results shows that recently proposed constant Q cepstral coefficients (CQCC) shows superior performance in comparison to standard Mel-frequency cepstral coefficients (MFCC).

Linear prediction residual features for automatic speaker verification anti-spoofing

Automatic speaker verification (ASV) systems are highly vulnerable against spoofing attacks. Anti-spoofing, determining whether a speech signal is natural/genuine or spoofed, is very important for improving the reliability of the ASV systems. Spoofing attacks using the speech signals generated using speech synthesis and voice conversion have recently received great interest due to the 2015 edition of Automatic Speaker Verification Spoofing and Countermeasures Challenge (ASVspoof 2015). In this paper, we propose to use linear prediction (LP) residual based features for anti-spoofing. Three different features extracted from LP residual signal were compared using the ASVspoof 2015 database. Experimental results indicate that LP residual phase cepstral coefficients (LPRPC) and LP residual Hilbert envelope cepstral coefficients (LPRHEC) obtained from the analytic signal of the LP residual yield promising results for anti-spoofing. The proposed features are found to outperform standard Mel-frequency cepstral coefficients (MFCC) and Cosine Phase (CosPhase) features. LPRPC and LPRHEC features give the smallest equal error rates (EER) for eight spoofing methods out of ten spoofing attacks in comparison to MFCC and CosPhase features.

Spectral Features for Synthetic Speech Detection

Recent advancements in voice conversion (VC) and speech synthesis research make speech-based biometric systems highly prone to spoofing attacks. This can provoke an increase in false acceptance rate in such systems and requires countermeasure to mitigate such spoofing attacks. In this paper, we first study the characteristics of synthetic speech vis-a-vis natural speech and then propose a set of novel short-term spectral features that can efficiently capture the discriminative information between them. The proposed features are computed using inverted frequency warping scale and overlapped block transformation of filter bank log energies. Our study presents a detailed analysis of antispoofing performance with respect to the variations in the warping scale for inverted frequency and block size for the block transform. For performance analysis, Gaussian mixture model (GMM) based synthetic speech detector is used as a classifier on a stand-alone basis and also, integrated with automatic speaker verification (ASV) systems. For ASV systems, standard mel-frequency cepstral coefficients are used as feature while GMM with universal background model and i-vector are used as classifiers. The experiments are conducted on ten different kinds of synthetic data from ASVspoof 2015 corpus. The results show that the countermeasures based on the proposed features outperform other spectral features for both known and unknown attacks. An average equal error rate (EER) of 0.00% has been achieved for nine attacks that use VC or SS speech and the best performance of 7.12% EER is arrived at the remaining natural speech concatenation-based spoofing attack.

Robust Accent Recognition in Malaysian English using PCA-Transformed Mel-Bands Spectral Energy Statistical Descriptors

The standard speech feature extractors such as Mel-Frequency Cepstral Coefficients (MFCC) and Linear Prediction Coefficients (LPC) fail to perform well under noisy conditions. In this paper two noise less-susceptible features are proposed to mitigate the deficiency of MFCC and LPC. Statistical descriptors of Mel-Bands Spectral Energy (MBSE) is applied to the traditional filter-bank analysis, however, this technique increases the feature size. This issue is tackled by proposing a transformation using principle component analysis to generate a new PCA-MBSE feature set. Two types of utterances namely isolated words and continuous speech were elicited from 103 university volunteers in a controlled room to collect speech signals from three main ethnic groups in Malaysia. This study employed two classifiers namely K-nearest neighbors and artificial neural networks to recognize between the Malay, Chinese and Indian accents. Experimental results using independent test samples technique indicated promising accuracy rates of 92.7% and 93.0% using the proposed PCA-MBSE features to recognize between the Malay, Chinese and Indian accents on the male and female datasets respectively. It was found that under severe noisy conditions, the standard MFCC and LPC features started to deteriorate faster than the MBSE-based features. PCA-MBSE features showed the most robust quality where its performance was just slightly deteriorated by 17.1% and 13.6% as compared to MBSE features i.e. 33.1% and 31.3% on the male and female datasets respectively. Further poor results of LPC features were obtained indicating deterioration rates of 40.2% and 32.7%, while that of MFCC features of 35.7% and 36.8% for the male and female datasets respectively. As a conclusion, Malaysian English is a not a uniform English variety colored by its diverse ethnic nuances. Incorporating accent analyzers using the proposed techniques in automatic speech recognition can contribute a substantial improvement in noisy environment.

Statistical Descriptors of Mel-Bands Spectral Energy Features with Feature Reduction for Robust Accent Recognition in Malaysian English

To date, Malaysian English (MalE) accents arises from different ethnics of its populace are scarcely investigated using empirical methods that give a decisive conclusion to treat MalE as either uniform or non-uniform variety. The popularly used Mel-Frequency Cepstral Coefficients (MFCC) and Linear Prediction Coefficients (LPC) as feature extractors fail to perform well under noisy conditions. This paper proposes two new methods and noise less-susceptible feature extractors to mitigate the deficiency of MFCC and LPC. Statistical descriptors of Mel-Bands Spectral Energy (MBSE) is an enhancement of traditional filter-bank analysis, however, increases fourfold as much the feature size. This issue is tackled by proposing a transformation using principle component analysis to generate a new PCA-MBSE feature set. Experimental results indicated promising accuracy rates of 92.7% and 93.0% using the proposed PCA-MBSE features to recognize between the Malay, Chinese and Indian accents of MalE speech for the male and female datasets respectively. It was found that under severe noisy conditions, the standard MFCC and LPC features started to deteriorate faster than the MBSE-based features. PCA-MBSE features showed the most robust quality where its performance was just slightly deteriorated by 17.1% and 13.6% as compared to MBSE features i.e. 33.1% and 31.3% for the male and female datasets respectively. Further poor results of LPC features were obtained indicating deterioration rates of 40.2% and 32.7%, while that of MFCC features of 35.7% and 36.8% for the male and female datasets respectively.

Recognizing the message and the messenger: biomimetic spectral analysis for robust speech and speaker recognition.

Humans are quite adept at communicating in presence of noise. However most speech processing systems, like automatic speech and speaker recognition systems, suffer from a significant drop in performance when speech signals are corrupted with unseen background distortions. The proposed work explores the use of a biologically-motivated multi-resolution spectral analysis for speech representation. This approach focuses on the information-rich spectral attributes of speech and presents an intricate yet computationally-efficient analysis of the speech signal by careful choice of model parameters. Further, the approach takes advantage of an information-theoretic analysis of the message and speaker dominant regions in the speech signal, and defines feature representations to address two diverse tasks such as speech and speaker recognition. The proposed analysis surpasses the standard Mel-Frequency Cepstral Coefficients (MFCC), and its enhanced variants (via mean subtraction, variance normalization and time sequence filtering) and yields significant improvements over a state-of-the-art noise robust feature scheme, on both speech and speaker recognition tasks.

SPARSE-BASED AUDITORY MODEL FOR ROBUST SPEAKER RECOGNITION

The mismatch between the training and the testing environments greatly degrades the performance of speaker recognition. Although many robust techniques have been proposed, speaker recognition in mismatch condition is still a challenge. To solve this problem, we propose a sparse-based auditory model as the front-end of speaker recognition by simulating auditory processing of speech signal. To this end, we introduce narrow-band filter-bank instead of the widely used wide-band filter-bank to simulate the basilar membrane filter-bank, use sparse representation as the approximation of basilar membrane coding strategy, and incorporate the frequency selectivity enhance mechanism between tectorial membrane and basilar membrane by practical engineering approximation. Compared with the standard Mel-frequency cepstral coefficient approach, our preliminary experimental results indicate that the sparse-based auditory model consistently improve the robustness of speaker recognition in mismatched condition.

Gender Effect Canonicalization for Bangla ASR

This paper presents a Bangla (widely used as Bengali) automatic speech recognition system (ASR) by suppressing gender effects. Gender characteristic plays an important role on the performance of ASR. If there is a suppression process that represses the decrease of differences in acoustic-likelihood among categories resulted from gender factors, a robust ASR system can be realized. In the proposed method, we have designed a new ASR incorporating the Local Features (LFs) instead of standard mel frequency cepstral coefficients (MFCCs) as an acoustic feature for Bangla by suppressing the gender effects, which embeds three HMM-based classifiers for corresponding male, female and geneder-independent (GI) characteristics. In the experiments on Bangla speech database prepared by us, the proposed system has achieved a significant improvement of word correct rates (WCRs), word accuracies (WAs) and sentence correct rates (SCRs) in comparison with the method that incorporates Standard MFCCs.

Model-Based Feature Enhancement for Reverberant Speech Recognition

In this paper, we present a new technique for automatic speech recognition (ASR) in reverberant environments. Our approach is aimed at the enhancement of the logarithmic Mel power spectrum, which is computed at an intermediate stage to obtain the widely used Mel frequency cepstral coefficients (MFCCs). Given the reverberant logarithmic Mel power spectral coefficients (LMPSCs), a minimum mean square error estimate of the clean LMPSCs is computed by carrying out Bayesian inference. We employ switching linear dynamical models as an a priori model for the dynamics of the clean LMPSCs. Further, we derive a stochastic observation model which relates the clean to the reverberant LMPSCs through a simplified model of the room impulse response (RIR). This model requires only two parameters, namely RIR energy and reverberation time, which can be estimated from the captured microphone signal. The performance of the proposed enhancement technique is studied on the AURORA5 database and compared to that of constrained maximum-likelihood linear regression (CMLLR). It is shown by experimental results that our approach significantly outperforms CMLLR and that up to 80% of the errors caused by the reverberation are recovered. In addition to the fact that the approach is compatible with the standard MFCC feature vectors, it leaves the ASR back-end unchanged. It is of moderate computational complexity and suitable for real time applications.

Feature extraction using Mel frequency cepstral coefficients for hyperspectral image classification

The Mel frequency cepstral coefficient (MFCC) model, which is widely used in speech detection and recognition, is introduced to extract features from hyperspectral image data. The similarities and differences between speech signals and spectral image data are compared and analyzed. The standard MFCC model is then improved to suit the characteristics of spectral image data by reintroducing the discarded phase information. Finally, the proposed model is applied to two real hyperspectral subimages. Experimental results show that the MFCC feature is sensitive and discriminative among reflectance spectra. It can be used as an effective feature extraction method for hyperspectral image classification.

Unsupervised Data-Driven Feature Vector Normalization With Acoustic Model Adaptation for Robust Speech Recognition

In this paper, an unsupervised data-driven robust speech recognition approach is proposed based on a joint feature vector normalization and acoustic model adaptation. Feature vector normalization reduces the acoustic mismatch between training and testing conditions by mapping the feature vectors towards the training space. Model adaptation modifies the parameters of the acoustic models to match the test space. However, since neither is optimal, both approaches use an intermediate space between training and testing spaces to map either the feature vectors or acoustic models. The joint optimization of both approaches provides a common intermediate space with a better match between normalized feature vectors and adapted acoustic models. In this paper, feature vector normalization is based on a minimum mean square error (MMSE) criterion. A class dependent multi-environment model linear normalization (CD-MEMLIN) based on two classes (silence/speech) with a cross probability model (CD-MEMLIN-CPM) is used. CD-MEMLIN-CPM assumes that each class of clean and noisy spaces can be modeled with a Gaussian mixture model (GMM), training a linear transformation for each pair of Gaussians in an unsupervised data-driven training process. This feature vector normalization maps the recognition space feature vector to a normalized space. The acoustic model adaptation maps the training space to the normalized space by defining a set of linear transformations over an expanded HMM-state space, compensating for those degradations that the feature vector normalization is not able to model, like rotations. Experiments have been carried out with the Spanish SpeechDat Car database and Aurora 2 databases using both the standard Mel-frequency cepstral coefficient (MFCC) and advanced ETSI front-ends. Consistent improvements were reached for both corpora and front-ends. Using the standard MFCC front-end, a 92.08% average improvement on WER for Spanish SpeechDat Car and a 69.75% average improvement for clean condition evaluation of Aurora 2 was obtained, improving those results reached with ETSI advanced front-end (83.28% and 67.41%, respectively). Using the ETSI advanced front-end with the proposed solution, a 75.47% average improvement was obtained for the clean condition evaluation of Aurora 2 database.

Robustness evaluation of wavelet based features for continuous speech recognition

This article evaluates robustness of admissible wavelet packet based features for continuous speech recognition. The recognition accuracy is compared with the standard mel frequency cepstral coefficients (MFCC) under clean and noisy environment. This is carried out by adding white Gaussian and speech noise to the phonemes of the TIMIT database to generate different levels of signal to noise ratio. Further, a wavelet based denoising technique is proposed as a front-end for noise reduction. Soft and hard thresholding techniques are used with one-level and two-level wavelet based denoising. The speech recogniser with Continuous Density Hidden Markov's Model is used to model the phonemes for the word recognition task. The recognition performance achieved with denoising of the input speech shows improvement as compared to without denoising of both the MFCC and the wavelet based features.

Automatic Recognition of Pathological Phoneme Production

Objective: Proper diagnosis and therapy of pathological pronunciation of phonemes play an important role in modern logopedics. To enhance the efficiency of diagnosis and therapy an automatic recognition of pathological phoneme pronunciation is addressed in this paper. The authors focus on the therapy of phoneme substitution disorders. Patients and Methods: Recognized speech samples come from speech-impaired Polish children and partially from persons imitating speech disorders. Recognized speech disorders were substitutions in pairs {s, ʂ}, {ɕ, ʂ}, {ʦ, tʂ+}, {ʨ, tʂ+}, {ʣ, dʐ+ }, and {ʥ, dʐ+ } embedded in Polish carrier words. In order to detect substitutions in the recognized words, recently proposed human factor cepstral coefficients (HFCC) have been implemented. Efficiency of the HFCC approach was compared to the application of standard mel-frequency cepstral coefficients (MFCC) as a feature vector. Both dynamic time warping (DTW), working on whole words or embedded phoneme patterns, and hidden Markov models (HMM) were used as classifiers. The HMM classifier was based on whole-word models as well as phoneme models. Results present a comparative analysis of DTW and HMM methods. Conclusions: The superiority of HFCC features over those of MFCC was demonstrated. Results obtained by DTW methods, mainly by modified phoneme-based DTW classifier, were slightly better in comparison with the HMM classifier. Results obtained for the detection of substitution in pairs {s, ʂ}, {ʦ, tʂ+}, {ʣ, ʣ+ } are very promising. The methods developed for these cases can be integrated into computer systems for speech therapy. For substitutions in pairs {ɕ, ʂ}, {ʨ, tʂ+}, {ʥ, dʐ+ } further research is necessary.

Low-dimensional, auditory feature vectors that improve vocal-tract-length normalization in automatic speech recognition

Human speech recognition is robust to large changes in vocal tract length (VTL) but automatic speech recognition is not. In an effort to improve VTL normalization, an auditory model was used to derive formant-like features from syllables. The robustness supported by these auditory features was compared to the robustness provided by traditional MFCCs (Mel-Frequency Cepstral Coefficients), using a standard HMM recognizer (Hidden-Markov-Model). The speech database consisted of 180 syllables, each scaled with the vocoder STRAIGHT to have a wide range VTLs and glottal pulse rates. Training took place with syllables from a small, central range of scale values. When tested on the full range of scaled syllables, average performance for MFCC-based recognition was 73.5%, with performance falling close to 0% for syllables with extreme VTL values. The feature vectors constructed with the auditory model led to much better performance; the average for the full range of scaled syllables was 91%, and performance never fell below 65% even for extreme combinations of VTL and GPR. Moreover the auditory feature vectors contain just 12 features whereas the standard MFCC vectors contain 39 features. Research supported by the UK-MRC (G0500221) and EOARD (FA8655-05-1-3043).