Feature Extraction Based on Pitch-Synchronous Averaging for Robust Speech Recognition

This paper presents an investigation of robust features for speech recognition in three different noisy environments. The state-of art Mel-frequency cepstral coefficients were extensively explored in additive, convolutive and reverberant environments. These environments have captured the interest of many researches in speech recognition systems. We evaluate robust speech recognition results on the TI-DIGIT database. Significant word error rate reductions were observed in the connected digit recognition experiments. The recognition experiments vindicate the robustness of Mel-frequency cepstral coefficient with dynamic features and cepstral mean normalization in hostile environments, especially additive and reverberant noise

Intra-speaker variations due to perceptually induced stress or emotion adversely affect speech recognition system performance. In this paper, we combine auditory based (Mel frequency cepstral coefficients and linear predictive cepstral coefficients) features and articulatory based (voicedness) features for robust speech recognition. Voicedness features are derived using linear and teager energy operator (TEO) based nonlinear fast Fourier transform (FFT) spectra. Nonlinear properties are analyzed in both the time and frequency domains. In addition, we investigate the sensitivity of all these FFT spectra to stress and observe the performance of individual FFT spectra. The system is tested using stressed speech data from the speech under simulated and actual stress (SUSAS) database. The results show that articulatory based features help to improve the system performance. Furthermore, significant performance improvement has been observed when using the FFT spectrum which is less sensitive to stress

Recognition of car horns based on principal component analysis of MEL frequency Cepstral coefficients and support vector machine

Combined speech enhancement and auditory modelling for robust distributed speech recognition

Speech recognition can be defined as the process of converting voice signals into the ranks of the word, by applying a specific algorithm that is implemented in a computer program. The research of speech recognition in Indonesia is relatively limited. This paper has studied methods of feature extraction which is the best among the Linear Predictive Coding (LPC) and Mel Frequency Cepstral Coefficients (MFCC) for speech recognition in Indonesian language. This is important because the method can produce a high accuracy for a particular language does not necessarily produce the same accuracy for other languages, considering every language has different characteristics. Thus this research hopefully can help further accelerate the use of automatic speech recognition for Indonesian language. There are two main processes in speech recognition, feature extraction and recognition. The method used for comparison feature extraction in this study is the LPC and MFCC, while the method of recognition using Hidden Markov Model (HMM). The test results showed that the LPC method is better than MFCC in Indonesian language speech recognition.

Recently, Li et al. proposed a new auditory feature for robust speech recognition in noise environments. The new feature was derived by mimicking closely the function of human auditory process. Several filters were used to model the outer ear, middle ear, and cochlea, and the initial filter parameters and shapes were obtained from crude psychoacoustics results, experience, or experiments. Although one may adjust the feature parameters by hand to get better performance, the resulting feature parameters still may not be optimal in the sense of minimal recognition errors, especially for different tasks. To further improve the auditory feature, in this paper we apply discriminative training to optimize the auditory feature parameters with some guidance from psychoacoustic evidence but otherwise in a data-driven approach so as to minimize the recognition errors. One significant contribution over similar efforts in the past, such as discriminative feature extraction, is that we make no assumption on the parametric form of the auditory filters. Instead, we only require the filters to be smooth and triangular-like as suggested by psychoacoustics research. Our approach is evaluated on the Aurora database and achieves a word error reduction of 19.2%.

Speaker identification accuracy decreases significantly in the presence of additive noise. In this paper, we propose a robust speech feature extraction method, which is based on the harmonic structure of voiced segments. The robust features are composed of fundamental and harmonic peak data from short-time spectrum. These features are evaluated by thirty speaker data from TIMIT database and additive noise signals from NOISEX-92 database with clean training and noisy testing samples. Results reflect that under low SNR (signal-to-noise ratio) environments new features achieve better performance than conventional MFCC (Mel-Frequency Cepstral Coefficients) parameters.

A method is proposed to improve the accuracy of the parameter estimates of highly damped sinusoidal signals in additive noise. The method first utilizes singular value decomposition (SVD) of the signal matrix to reduce the noise effect, then based on the noise-reduced signals the weighted sum of squared errors is minimized using an adaptive lattice filter algorithm to estimate the parameters in the backward linear prediction model. In the case of large damping factors, the experimental results show that the proposed method gives more accurate estimates than the eigenvector method. >

Prior studies in the automatic classification of voice quality have mainly studied the use of the acoustic speech signal as input. Recently, a few studies have been carried out by jointly using both speech and neck surface accelerometer (NSA) signals as inputs, and by extracting mel-frequency cepstral coefficients (MFCCs) and glottal source features. This study examines simultaneously-recorded speech and NSA signals in the classification of voice quality (breathy, modal, and pressed) using features derived from three self-supervised pre-trained models (wav2vec2-BASE, wav2vec2-LARGE, and HuBERT) and using a support vector machine (SVM) as well as convolutional neural networks (CNNs) as classifiers. Furthermore, the effectiveness of the pre-trained models is compared in feature extraction between glottal source waveforms and raw signal waveforms for both speech and NSA inputs. Using two signal processing methods (quasi-closed phase (QCP) glottal inverse filtering and zero frequency filtering (ZFF)), glottal source waveforms are estimated from both speech and NSA signals. The study has three main goals: (1) to study whether features derived from pre-trained models improve classification accuracy compared to conventional features (spectrogram, mel-spectrogram, MFCCs, i-vector, and x-vector), (2) to investigate which of the two modalities (speech vs. NSA) is more effective as input in the classification task with pre-trained model-based features, and (3) to evaluate whether the deep learning-based CNN classifier can enhance the classification accuracy in comparison to the SVM classifier. The results revealed that the use of the NSA input showed better classification performance compared to the speech signal. Between the features, the pre-trained model-based features showed better classification accuracies, both for speech and NSA inputs compared to the conventional features. The two classifiers performed equally well for all the pre-trained model-based features for both speech and NSA signals. It was also found that the HuBERT features performed better than the wav2vec2-BASE and wav2vec2-LARGE features for both speech and NSA inputs. In particular, when compared to the conventional features, the HuBERT features showed an absolute accuracy improvement of 3%–6% for speech and NSA signals in the classification of voice quality.

We make use of information inside infant’s cry signal in order to identify the infant’s psychological condition. Gaussian mixture models (GMMs) are applied to distinguish between healthy full-term and premature infants, and those with specific medical problems available in our cry database. Cry pattern for each pathological condition is created by using adapted boosting mixture learning (BML) method to estimate mixture model parameters. In the first experiment, test results demonstrate that the introduced adapted BML method for learning of GMMs has a better performance than conventional EM-based reestimation algorithm as a reference system in multipathological classification task. This newborn cry-based diagnostic system (NCDS) extracted Mel-frequency cepstral coefficients (MFCCs) as a feature vector for cry patterns of newborn infants. In binary classification experiment, the system discriminated a test infant’s cry signal into one of two groups, namely, healthy and pathological based on MFCCs. The binary classifier achieved a true positive rate of 80.77% and a true negative rate of 86.96% which show the ability of the system to correctly identify healthy and diseased infants, respectively.

Sound feature extraction Mel Frequency Cepstral Coefficients (MFCC) and Vector Quantization (VQ) classification Linde-Buzo-Gray algorithm (LBG) algorithms are applied for recognizing the background sounds in the human daily activities. Applying these algorithms to twenty typical daily activity sounds, average recognition accuracy of 93.8% can be achieved. In these algorithms, how three parameters (i.e., Mel filters number, frame-to-frame overlap and LBG codebook cluster number) affect system's calculation burden and accuracy is also investigated. By adjusting these three parameters to an optimized combination, the multiplication and addition calculation burden can be reduced by 87.0% and 87.1% individually while maintaining the system's average accuracy rate at 92.5%. This is promising for future integration with other sensor (s) to fulfill daily activity recognition by using power aware Wireless Sensor Networks (WSN) systems.

Respiratory disease presents major global health challenges, and its early detection could save numerous lives, prompting the demand for accurate and non-intrusive diagnostic methods. This study presents a multi-model machine learning framework for classifying respiratory diseases (COPD, Bronchiolitis, Pneumonia, URTI, and Healthy) using audio data from the Respiratory Sound Database. We extract Mel frequency cepstral coefficients (MFCCs) and apply data augmentation to address class imbalance. We propose a novel hybrid model, ConvMambda, and compare its performance against other popular classification models, including 1D CNN, random forest, SVM, and LSTM. The ConvMamba model achieves the highest test accuracy (90.77%) and macro F1-score (0.8932), outperforming 1D-CNN (90% accuracy) and random forest (82% accuracy), SVM (68% accuracy), and LSTM (68% accuracy). We analyze the performance across classes, highlighting challenges with minority classes (Bronchiolitis, URTI) and propose strategies to mitigate imbalance. Our findings demonstrate the potential of audio-based diagnostics, with the ConvMamba model offering the superior performance for respiratory diseases classification.

Most current speech recognition systems are built upon a single type of model, e.g. an HMM or certain type of segment based model, and furthermore typically employs only one type of acoustic feature e.g. MFCCs and their variants. This entails that the system may not be robust should the modeling assumptions be violated. Recent research efforts have investigated the use of multi-scale/multi-band acoustic features for robust speech recognition. This paper described a multi-model approach as an alternative and complement to the multi-feature approaches. The multi-model approach seeks a combination of different types of acoustic models, thereby integrating the capabilities of each individual model for capturing discriminative information. An example system built upon the combination of the standard HMM technique with a segment-based modeling technique was implemented. Experiments for both isolated-word and continuous speech recognition have shown improved performances over each of the individual models considered in isolation.

Exploring joint equalization of spatial-temporal contextual statistics of speech features for robust speech recognition

This paper proposes a neural mechanism motivated system to extract noise resistant features for robust speech recognition. We use nonnegative matrix factorization to construct two layers of auditory neurons which captures the essence of speech patterns. The responses of these neurons to speech are further processed to form an auditory neural cepstral coefficient (ANCC) representation for speech recognition. We test the robustness of ANCC feature on a 51-word corpus, with recognizers trained on clean speech in noisy conditions. Compared with MFCC, ANCC shows less performance degradation and achieves satisfactory recognition accuracies in both non-stationary noise and high noise level conditions.