Mel-frequency Cepstral Coefficients Vectors Research Articles

A computer-aided-diagnosis system for neuromuscular diseases using Mel frequency Cepstral coefficients

Amyotrophic Lateral Sclerosis (ALS) and Myopathy are the most well-known neuromuscular diseases. Electromyography (EMG) signal is hugely used in the diagnosis of these neuromuscular disorders. The study presented in this paper aims to develop a Computer-Aider-Diagnosis (CAD) system to identify ALS and Myopathic patients from Normal subjects through the EMG signal. The system employs the Mel Frequency Cepstral Coefficients (MFCC) as a feature extraction technique to produce discriminant features. Then, the MFCC vectors dimension is reduced using statistical values. Since features constitute a crucial factor in characterizing several phenomena and reaching good performances by CAD systems, the Relief feature selection algorithm was applied to the previously obtained outputs to select the most meaningful and significant attributes. The main contribution of this work is to build a robust model for the classification of EMG signals to identify the previously mentioned neuromuscular disorders. Hence, the last stage of the proposed scheme deals with machine learning classification algorithms. Support Vector Machine (SVM) and k-Nearest Neighbors (k-NN) were chosen in this work. Experimental results showed that k-NN with two nearest neighbors (k=2) and the 10-folds cross-validation method yielded higher classification results than SVM. According to the accuracy, sensitivity, specificity, and F-measure evaluation metrics, the classifiers' performances were evaluated. In Normal-ALS and Normal-Myopathy binary issues, we obtained 99.34% and 99.07% accuracy, respectively. The proposed system performed well also in a multi-class task with a classification accuracy of 98.69%.

Coarticulation varies by consonant identity in adult and child speech

Consonant-vowel coarticulation varies by manner of articulation: speakers coarticulate less between segments in V[stop] sequences than V[glide]. There are different approaches to modelling consonant identity effects upon coarticulation (Iskarous et al., 2013). However, these effects have not been studied in non-literate populations—young children or non-literate adults. This is relevant since literacy heightens phoneme awareness, affecting coarticulation. 40 South Bolivian Quechua speakers who were not literate in the language, 30 children (aged 5–10) and 10 adults, completed a picture-prompted word elicitation task. Coarticulation was measured between [a] and the consonants /p, t, k, kh, qh, ch, s, m, w/ in VC and CV syllables. Because the child speech apparatus poses problems for typical spectral analysis, we quantify coarticulation as the Euclidean distance between Mel frequency cepstral coefficient vectors averaged over adjacent phones (Gerosa et al., 2006). Mixed effects models predicting the Euclidean distance between adjacent phones show that all age groups are sensitive to changes in consonant manner. Adults and children restrained coarticulation more between sequences containing obstruents than glides or liquids. These results suggest that literacy and heightened phoneme awareness are not required for consonant identity to affect coarticulation.Consonant-vowel coarticulation varies by manner of articulation: speakers coarticulate less between segments in V[stop] sequences than V[glide]. There are different approaches to modelling consonant identity effects upon coarticulation (Iskarous et al., 2013). However, these effects have not been studied in non-literate populations—young children or non-literate adults. This is relevant since literacy heightens phoneme awareness, affecting coarticulation. 40 South Bolivian Quechua speakers who were not literate in the language, 30 children (aged 5–10) and 10 adults, completed a picture-prompted word elicitation task. Coarticulation was measured between [a] and the consonants /p, t, k, kh, qh, ch, s, m, w/ in VC and CV syllables. Because the child speech apparatus poses problems for typical spectral analysis, we quantify coarticulation as the Euclidean distance between Mel frequency cepstral coefficient vectors averaged over adjacent phones (Gerosa et al., 2006). Mixed effects models predicting the Euclide...

Speech Magnitude Spectrum Reconstruction from MFCCs Using Deep Neural Network

This work proposes a Deep neural network (DNN) based method for reconstructing speech magnitude spectrum from Mel-frequency cepstral coefficients (MFCCs). We train a DNN using MFCC vectors as input and the corresponding speech magnitude spectrum as desired output. Exploiting the strong inference power of DNN, the proposed method has the capability to accurately estimate the speech magnitude spectrum even from truncated MFCC vectors. Experiments on TIMIT corpus demonstrate that the proposed method achieves significantly better performance compared with traditional methods.

Holistic Lexical Storage: Coarticulatory Evidence from Child Speech

Adult speakers readily decompose morphologically-complex words into their component parts. Overgeneralizations in children’s early speech (e.g. goed) demonstrate that they must share in this ability. However, acoustic evidence from child speech suggests that children do not always break words down, instead storing language in more holistic chunks such as syllables or even entire words.How are morphologically-complex forms represented throughout childhood? To answer this, we measured coarticulation within and between morphemes in adult and child (age 5-10) South Bolivian Quechua speakers. Coarticulation was quantified as the difference between the averaged Mel frequency cepstral coefficient vectors of the adjacent phones. Experiment 1 replicates known coarticulatory findings from the literature, demonstrating the validity of the MFCC measurement for calculating adjacent coarticulation. Experiment 2 then measures coarticulation between a single biphone sequence, [ap], in two environments: 1) within morphemes and 2) across morpheme boundaries. Results show that adult speakers coarticulated less across morpheme boundaries than within root morphemes. This is further evidence that adults decompose complex words. Children, however, coarticulated equally across and within morphemes. This suggests that the child speakers store inflected words more holistically than adults, even in this highlyagglutinating language.

Perceptually Weighted Analysis-by-Synthesis Vector Quantization for Low Bit Rate MFCC Codec

This letter presents a perceptually weighted analysis-by-synthesis vector quantization (VQ) algorithm for low bit rate MFCC codec. Different from conventional VQ of mel-frequency cepstral coefficients (MFCCs) vector, this algorithm uses an analysis-by-synthesis technique and aims to minimize the perceptually weighted spectral reconstruction distortion rather than the distortion of MFCCs vector itself. Also, to reduce the computational complexity, we propose a practical suboptimal codebook searching technique and embed it into the split and multistage VQ framework. Objective and subjective experimental results on Mandarin speech show that the proposed algorithm yields intelligible and natural sounding speech for speech coding at 600–2400 bit/s. Compared to current VQ in MFCC codec, the output speech quality is substantially improved in terms of frequency-weighted segmental SNR, short-time objective intelligibility score, perceptual evaluation of speech quality score, and mean opinion score.

IJARCCE - Computer and Communication Engineering

3 Abstract: A K-Nearest Neighbour Algorithm involving Mel-Frequency Cepstral Coefficients (MFCCs) is provided to perform Speech signal feature extraction for the task of speaker accent recognition. Mel-Frequency Cepstral Co- efficient is effectively used to perform the feature extraction of the input signal. For each input signal the mean of the MFCC matrix is used for pattern recognition .The K-nearest neighbour algorithm is based on evaluating minimum Euclidean distance measure from input data set to stored data set. Since large number of speakers of different accent are present, they can be grouped together depending upon their accent .Thus each signal coming from different group makes a distinct MFCC vector .In this paper we have compared the MFCC from global group to smaller sub groups.

Burst Onset Landmark Detection and Its Application to Speech Recognition

The reliable detection of salient acoustic-phonetic cues in speech signal plays an important role in speech recognition based on speech landmarks. Once speech landmarks are located, not only can phone recognition be performed, but other useful information can also be derived. This paper focuses on the detection of burst onset landmarks, which are crucial to the recognition of stop and affricate consonants. The proposed detector is purely based on a random forest technique, which belongs to an ensemble of tree-structured classifiers. By adopting a special asymmetric bootstrapping method, a series of experiments conducted on the TIMIT database demonstrate that the proposed detector is an efficient and accurate method for detecting burst onsets. When the detection results are appended to mel frequency cepstral coefficient vectors, the augmented feature vectors enhance the recognition correctness of hidden Markov models in recognizing stop and affricate consonants in continuous speech.

A new approach for small-signal modelling of the field effect transistor based on cepstral coefficients and discrete transforms

In this article, a new technique for field effect transistor (FET) small-signal modelling using neural networks is proposed. This technique is based on the combination of the Mel-frequency cepstral coefficients (MFCCs) with different discrete transforms such as the discrete cosine transform (DCT), the discrete sine transform (DST) and the discrete wavelet transform (DWT) of the inputs to the neural networks. The input datasets to traditional neural systems for FET small-signal modelling are the scattering parameters and the corresponding frequencies in a certain band, and the output datasets are the circuit elements. In the proposed approach, these datasets are considered to form random signals. The MFCCs of the random signals are used to generate a small number of features characterising the signals. In addition, other vectors are calculated from the DCT, DST or DWT of the random signals and appended to the MFCC vectors calculated from the signals. The new feature vectors are used to train the neural networks. The objective of using these new vectors is to characterise the random input sequences with more features, to encourage robustness against measurement errors. There are two benefits to these approaches: (1) a reduction in the number of neural network inputs, and hence a faster convergence of the neural training algorithm and (2) robustness against measurement errors in the testing phase. Experimental results show that the techniques based on the discrete transforms are less sensitive to measurement errors than using the traditional and MFCC methods.

Robust Acoustic Speech Feature Prediction From Noisy Mel-Frequency Cepstral Coefficients

This paper examines the effect of applying noise compensation to acoustic speech feature prediction from noisy mel-frequency cepstral coefficient (MFCC) vectors within a distributed speech recognition architecture. An acoustic speech feature (comprising fundamental frequency, formant frequencies, speech/nonspeech classification, and voicing classification) is predicted from an MFCC vector in a maximum a posteriori (MAP) framework using phoneme-specific or global models of speech. The effect of noise is considered and three different noise compensation methods, that have been successful in robust speech recognition, are integrated within the MAP framework. Experiments show that noise compensation can be applied successfully to prediction with best performance given by a model adaptation method that performs only slightly worse than matched training and testing. Further experiments consider application of the predicted acoustic features to speech reconstruction. A series of human listening tests show that the predicted features are sufficient for speech reconstruction and that noise compensation improves speech quality in noisy conditions.

Analysis and prediction of acoustic speech features from mel-frequency cepstral coefficients in distributed speech recognition architectures

The aim of this work is to develop methods that enable acoustic speech features to be predicted from mel-frequency cepstral coefficient (MFCC) vectors as may be encountered in distributed speech recognition architectures. The work begins with a detailed analysis of the multiple correlation between acoustic speech features and MFCC vectors. This confirms the existence of correlation, which is found to be higher when measured within specific phonemes rather than globally across all speech sounds. The correlation analysis leads to the development of a statistical method of predicting acoustic speech features from MFCC vectors that utilizes a network of hidden Markov models (HMMs) to localize prediction to specific phonemes. Within each HMM, the joint density of acoustic features and MFCC vectors is modeled and used to make a maximum a posteriori prediction. Experimental results are presented across a range of conditions, such as with speaker-dependent, gender-dependent, and gender-independent constraints, and these show that acoustic speech features can be predicted from MFCC vectors with good accuracy. A comparison is also made against an alternative scheme that substitutes the higher-order MFCCs with acoustic features for transmission. This delivers accurate acoustic features but at the expense of a significant reduction in speech recognition accuracy.

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).

Prediction of Fundamental Frequency and Voicing From Mel-Frequency Cepstral Coefficients for Unconstrained Speech Reconstruction

This work proposes a method for predicting the fundamental frequency and voicing of a frame of speech from its mel-frequency cepstral coefficient (MFCC) vector representation. This information is subsequently used to enable a speech signal to be reconstructed solely from a stream of MFCC vectors and has particular application in distributed speech recognition systems. Prediction is achieved by modeling the joint density of fundamental frequency and MFCCs. This is first modeled using a Gaussian mixture model (GMM) and then extended by using a set of hidden Markov models to link together a series of state-dependent GMMs. Prediction accuracy is measured on unconstrained speech input for both a speaker-dependent system and a speaker-independent system. A fundamental frequency prediction error of 3.06% is obtained on the speaker-dependent system in comparison to 8.27% on the speaker-independent system. On the speaker-dependent system 5.22% of frames have voicing errors compared to 8.82% on the speaker-independent system. Spectrogram analysis of reconstructed speech shows that highly intelligible speech is produced with the quality of the speaker-dependent speech being slightly higher owing to the more accurate fundamental frequency and voicing predictions

MAP prediction of formant frequencies and voicing class from MFCC vectors in noise

Novel methods are presented for predicting formant frequencies and voicing class from mel-frequency cepstral coefficients (MFCCs). It is shown how Gaussian mixture models (GMMs) can be used to model the relationship between formant frequencies and MFCCs. Using such models and an input MFCC vector, a maximum a posteriori (MAP) prediction of formant frequencies can be made. The specific relationship each speech sound has between MFCCs and formant frequencies is exploited by using state-specific GMMs within a framework of a set of hidden Markov models (HMMs). Formant prediction accuracy and voicing prediction of speaker-independent male speech are evaluated on both a constrained vocabulary connected digits database and a large vocabulary database. Experimental results show that for HMM–GMM prediction on the connected digits database, voicing class prediction error is less than 3.5%. Less than 1.8% of frames have formant frequency percentage errors greater than 20% and the mean percentage error of the remaining frames is less than 3.7%. Further experiments show prediction accuracy under noisy conditions. For example, at a signal-to-noise ratio (SNR) of 0 dB, voicing class prediction error increases to 9.4%, less than 4.3% of frames have formant frequency percentage errors over 20% and the formant frequency percentage error for the remaining frames is less than 5.7%.

Predicting fundamental frequency from mel-frequency cepstral coefficients to enable speech reconstruction

This work proposes a method to reconstruct an acoustic speech signal solely from a stream of mel-frequency cepstral coefficients (MFCCs) as may be encountered in a distributed speech recognition (DSR) system. Previous methods for speech reconstruction have required, in addition to the MFCC vectors, fundamental frequency and voicing components. In this work the voicing classification and fundamental frequency are predicted from the MFCC vectors themselves using two maximum a posteriori (MAP) methods. The first method enables fundamental frequency prediction by modeling the joint density of MFCCs and fundamental frequency using a single Gaussian mixture model (GMM). The second scheme uses a set of hidden Markov models (HMMs) to link together a set of state-dependent GMMs, which enables a more localized modeling of the joint density of MFCCs and fundamental frequency. Experimental results on speaker-independent male and female speech show that accurate voicing classification and fundamental frequency prediction is attained when compared to hand-corrected reference fundamental frequency measurements. The use of the predicted fundamental frequency and voicing for speech reconstruction is shown to give very similar speech quality to that obtained using the reference fundamental frequency and voicing.