Shouted speech detection using hidden markov model with rahmonic and mel-frequency cepstrum coefficients

The model-based speech enhancement method usually models types of noises in a prior and selects one noise model in the enhancement phase. In this paper, we study the modeling and selection (i.e. noise identification) algorithm for the model-base speech enhancement. For the noise signal features, Mel Frequency Cepstrum Coefficient (MFCC), Liner Prediction Coefficient (LPC), Linear Spectral Frequency (LSF) and Linear Predictive Cepstrum Coefficients (LPCC) are utilized. For the noise model, codebook, Gaussian Mixture Model (GMM) and Hidden Markov Model (HMM) are adopted. We couple these features and models and provide a comparative performance analysis of them. In the experiment, we use NOIZEUS database for noise modeling and testing. The experiment results show that MFCC and GMM is an excellent couple for noise identification, and LFS can take the place of LPC in the codebook-based speech enhancement.

A hybrid model comprised of Gaussian Mixtures Models (GMMs) and Hidden Markov Models (HMMs) is used to model generic sounds with large intra class perceptual variations. Each class has variable number of mixture components in the GMM. The number of mixture components is derived using the Minimum Description Length (MDL) criterion. The overall performance of the hybrid model was compared against models based on HMMs and GMMs with a fixed number of mixture components across all classes. We show that a hybrid model outperforms both class-based GMMs, HMMs, and GMMs based on fixed number of components. Further, our experiments revealed that the contribution of transitions between states in HMMs has no significant effect on the overall classification performance of generic sounds when large intra class perceptual variations are present among sounds in the training and test datasets. Sounds that show multi-event structure with events that tend to be similar (repetitive) indicated improved performance when modeled with HMMs that can be attributed to HMM’s state transition property. Conversely, GMMs indicate better performance when the sound samples show subtle or no repetitive behavior. These results were validated using the MuscleFish sound database.

In this paper, based on the study of two speech recognition algorithms, two designs of speech recognition system are given to realize this isolated speech recognition mobile robot control system based on ARM9 processor. The speech recognition process includes pretreatment of speech signal, characteristic extrication, pattern matching and post-processing. Mel-Frequency cepstrum coefficients (MFCC) and linear prediction cepstrum coefficients (LPCC) are the two most common parameters. Through analysis and comparison the parameters, MFCC shows more noise immunity than LPCC, so MFCC is selected as the characteristic parameters. Both dynamic time warping (DTW) and hidden markov model (HMM) are commonly used algorithm. For the different characteristics of DTW and HMM recognition algorithm, two different programs were designed for mobile robot control system. The effect and speed of the two speech recognition system were analyzed and compared.

Automatic Speech Emotion Recognition (SER) is a current research topic in the field of Human Computer Interaction (HCI) with a wide range of applications. The purpose of speech emotion recognition system is to automatically classify speaker's utterances into different emotional states such as disgust, boredom, sadness, neutral, and happiness. The speech samples in this paper are from the Berlin emotional database. Mel Frequency cepstrum coefficients (MFCC), Linear prediction coefficients (LPC), linear prediction cepstrum coefficients (LPCC), Perceptual Linear Prediction (PLP) and Relative Spectral Perceptual Linear Prediction (Rasta-PLP) features are used to characterize the emotional utterances using a combination between Gaussian mixture models (GMM) and Support Vector Machines (SVM) based on the Kullback-Leibler Divergence Kernel. In this study, the effect of feature type and its dimension are comparatively investigated. The best results are obtained with 12-coefficient MFCC. Utilizing the proposed features a recognition rate of 84% has been achieved which is close to the performance of humans on this database.

<p>In Ethiopia, the largest ethnic and linguistic groups are the Oromos, Amharas and Tigrayans. This paper presents the performance analysis of text-independent speaker identification system for the Amharic language in noisy environments. VQ (Vector Quantization), GMM (Gaussian Mixture Models), BPNN (Back propagation neural network), MFCC (Mel-frequency cepstrum coefficients), GFCC (Gammatone Frequency Cepstral Coefficients), and a hybrid approach had been use as techniques for identifying speakers of Amharic language in noisy environments. For the identification process, speech signals are collected from different speakers including both sexes; for our data set, a total of 90 speakers’ speech samples were collected, and each speech have 10 seconds duration from each individual. From these speakers, 59.2%, 70.9% and 84.7% accuracy are achieved when VQ, GMM and BPNN are used on the combined feature vector of MFCC and GFCC. </p>

This paper describes the development of an efficient speech recognition system using different techniques such as Mel Frequency Cepstrum Coefficients (MFCC), Vector Quantization (VQ) and Hidden Markov Model (HMM).This paper explains how speaker recognition followed by speech recognition is used to recognize the speech faster, efficiently and accurately.MFCC is used to extract the characteristics from the input speech signal with respect to a particular word uttered by a particular speaker.Then HMM is used on Quantized feature vectors to identify the word by evaluating the maximum log likelihood values for the spoken word.

Oil and gas resources pipelines, boundary lines and other places need to monitor their safety status in real time. The fiber early warning system becomes a good choice for its high sensitivity, corrosion resistance and concealment. The system provides early warning of the detection of fiber vibration signals. In this paper, an improved Mel frequency cepstrum coefficient (MFCC) method is proposed for the cepstrum characteristics recognition of different typical optical fiber vibration signals. Firstly, we pre-process the intrusion signals and obtain its power spectral density (PSD) to quantify the difference of frequency spectrum in respective intrusions. Secondly, the adaptive filter bank is designed according to the distribution of signal power spectrum to improve the conventional MFCC method. Through the analysis of the characteristic parameters, the MFCC coefficients are obtained. Finally, the Mean-crossing rates (MCR) of MFCC are calculated and the appropriate thresholds are selected to classify the typical vibration signals. Compared with the traditional MFCC, this improved MFCC method realizes adaptive division of frequency band according to the distribution of signal power spectrum. Experiments show that the algorithm can identify the manual signal, the mechanical signal and the vehicle signal in the research of the vibration signal recognition of the optical fiber pre-warning system (OFPS).

In this dissertation, a novel approach of integrating template matching with statistical modeling is proposed to improve continuous speech recognition. Hidden Markov Modeling (HMMs) has been the dominant approach in statistical speech recognition since it provides a principled way of jointly modeling speech spectral variations and time dynamics. However, HMMs have the shortcoming of assuming the observations being independent within each state, which makes it ineffective in modeling the details of speech temporal evolutions that are important for characterizing nonstationary speech sounds. Template-based methods make comparisons between a test pattern and the templates derived from training data, and therefore they are able to capture speech dynamics and time correlation of speech frames better than HMM based methods. However, template matching requires large memory space and computational time since feature vectors of training data need to be stored in computer memory for access at the recognition stage, which is difficult in large vocabulary continuous speech recognition (LVCSR). Our proposed approach takes advantages of both statistical modeling and template matching, which overcomes the weakness of conventional template-based method and is feasible for LVCSR. We use multiple Gaussian Mixture Model (GMM) indices to represent each frame of speech templates, and define the template unit to be context-dependent phone segments (triphone context). We also use phonetic decision trees borrowed from those commonly used in HMMs to tie triphone templates and predict triphones unseen in training data. Two local distances, log likelihood ratio (LLR) and Kullback-Leibler (KL) divergence, are proposed for dynamic time warping (DTW) based template matching. In order to reduce computational complexity and storage space, we propose methods of minimum distance template selection (MDTS) and maximum log-likelihood template selection (MLTS), and investigate a template compression method on top of template selection to further improve recognition performance. The template based methods were used to rescore lattices generated by baseline HMMs on the tasks of TIMIT continuous phone recognition and teleheath LVCSR and experimental results demonstrated that the proposed approach of integrating template matching with statistical modeling significantly improved recognition performances over the HMM baselines. The template selection methods also provided significant recognition accuracy improvements over the HMM baseline while largely reducing the computation and storage complexities. When all templates or MDTS were used, using the LLR local distance obtained better recognition performance than the KL divergence local distance. For MLTS and template compression, KL divergence local distance provided better performance than the LLR local distance, and the template compression method made further improvements over KL based MLTS. Since the templates were constructed based on the GMM indices extracted from HMM baselines, we also validated the effectiveness of the proposed template methods based on enhanced HMM baselines. Experimental results showed that LLR based all template method was able to consistently improve TIMIT phone recognition accuracies based on four enhanced HMM baselines. Prosodic features such as duration, energy, and pitch can reflect longer span information of speech than conventional single frame vectors but they have commonly been ignored by HMMs. Template based methods provide possibilities to conveniently integrate prosodic features into speech recognition, which has not been well studied in the past. In this dissertation, we investigate combining template based methods with the speech prosodic features of duration, energy and pitch to further improve speech recognition accuracy. The scores of prosodic information were computed by a GMM based method and a non-parametric method, and the prosodic scores were combined with the acoustic scores in triphone template matching. Experimental results obtained on the telehealth task showed that prosodic information had positive effects on vowel sound recognition.

Biometric information that exists in humans is unique from one human to another. One of the biometric data that is easily obtained is the human voice. The human voice is identic data that can differentiate between individuals. When we hear human voices directly, it is easy for our ears to tell the person who is speaking is male or female. But sometimes male voices can resemble girls and vice versa. Therefore, we propose a human voice detection system through Artificial Intelligence (AI) in machine learning. In this study, we used the Mel Frequency Cepstrum Coefficients (MFCC) method to extract human voice features and Gaussian Mixture Models (GMM) for the classification of female or male voice data. The experiment results showed that the system built was able to detect human gender through biometric voice data with an accuracy of 81.18%.

Gesture recognition is a quite promising field in robotics and many Human-Computer Interaction (HCI) related areas. This research uses Microsoft® Kinect to capture the 3D position data of joints, and uses Gaussian Mixture Model (GMM) and Hidden Markov Model (HMM) to model full-body gestures. We propose a gesture recognition algorithm to segment gestures from real-time data flow, and finally achieved to recognize predefined full-body gestures in real-time. This proposed method gives a high recognition rate of 94.36%, indicating the capability of the new method.

Now a day’s security and identification of a person are crucial in real time applications and solving the problems in biometric identification. Objectives: To address the issues in real time, the face and its organ’s identification are the main parts of the human body. The proposed work involves extraction of the face of the real time captured image, after extraction of the face, the organs like eyes, mouth and nose is extracted for identification of exact person. Methods/ Analysis: The present work captures the real time image from cameras or moving video devices for facial recognition using Principal Component Analysis (PCA), the facial features are extracted from both Hidden Markov Model (HMM), Gaussian mixture model (GMM) methods and classified into different organs using Artificial Neural Network (ANN). The features of organs are extracted in different stages, in first stage Eigen values with the help of PCA, second and third stage feature extraction with help GMM and HMM since these three techniques are the most powerful tools for statistical natural image processing. After extraction of organs of the face, the ANN is applied for classifications of eyes, mouth and nose separately. Findings: In the proposed work the facialorgans are separated into threeslight scale images and these are recombined to acquire the appreciationfacial image results like mouth, eyes and nose. The final obtained results shows that the proposed method has been achieved 95.8% recognition accuracy, Fault Rejection Ratio (FRR) is about 93.1% and Fault Acceptance Ratio (FAR) is 1.7 % which are implemented in Matlab2013A.

An audio-visual speech recognition system (AVSR) integrates audio and visual information to perform speech recognition task. The AVSR has various applications in practice especially in natural language processing systems such as speech-to-text conversion, automatic translation and sentiment analysis. Decades ago, researchers tend to use Hidden Markov Model (HMM) to construct speech recognition system due to its good achievements in success recognition rate. However, HMM's training dataset is enormous in order to have sufficient linguistic coverage. Besides, its recognition rate under noisy environments is not satisfying. To overcome this deficiency, a Recurrent Neural Network (RNN) based AVSR is proposed. The proposed AVSR model consists of three components: 1) audio features extraction mechanism, 2) visual features extraction mechanism and 3) audio and visual features integration mechanism. The features integration mechanism combines the output features from both audio and visual extraction mechanisms to generate final classification results. In this research, the audio features mechanism is modelled by Mel-frequency Cepstrum Coefficient (MFCC) and further processed by RNN system, whereas the visual features mechanism is modelled by Haar-Cascade Detection with OpenCV and again, it is further processed by RNN system. Then, both of these extracted features were integrated by multimodal RNN-based features-integration mechanism. The performance in terms of the speech recognition rate and the robustness of the proposed AVSR system were evaluated using speech under clean environment and Signal Noise Ratio (SNR) levels ranging from -20 dB to 20 dB with 5 dB interval. On average, final speech recognition rate is 89% across different levels of SNR.

In this paper, we propose a new method for voice feature extraction by using a hierarchical clustering approach of inverse discrete cosine transform (IDCT) cepstrum coefficient. Since the IDCT cepstrum coefficient is transformed into the feature vector based on Euclidean metric, we call the proposed voice feature as “E-vector”. Comparing with other voice features, e.g. Mel frequency cepstrum coefficient (MFCC) and histogram of DCT cepstrum coefficients (HDCC), the E-vector can represent the dynamic characteristics and interactional relationships of the voice. In the numerical experiments, we use the speech files based on the voice of 630 people in TIMIT corpus, and employ Gaussian mixture model to compare the recognition accuracy of E-vector with that of MFCC and HDCC. The results show that E-vector outperformed other voice features on personal identification, and presented higher extensiveness on voice feature extraction.

When utilizing the most likely state sequence (MLSS) criterion in Gauss mixture model-hidden Markov model (GMM-HMM) to acquire the best state series of observations, only the maximum likelihood state of speech frame is considered. Therefore, the influence of other states is neglected, which leads to the losing of some important information, and further reduces the recognition rate of the system. In this paper, we propose two new features, which are called state likelihood cluster feature (SLCF) and supervised state feature (SSF), to both reflect acoustic features and fuse state information. Combining SLCF and SSF with Mel frequency cepstrum coefficient (MFCC), Mel frequency cepstrum & state likelihood cluster feature (MSLCF) and Mel frequency cepstrum & supervised state feature (MSSF) are formed, respectively. By the proposed MSLCF and MSSF in Chinese speech recognition experiment, the relative error rate of the isolated word recognition system declines 6.10% and 9.66%, respectively, and the relative error rate of the continuous speech recognition system declines 2.53% and 11.05%, respectively.

With the wide application of electronic computers and artificial intelligent machines, people find that the best way of communication between man and machine is voice communication. Nowadays, highly accurate speaker identification systems are required. Speaker-specific characteristics exist in speech signals due to different speakers having different resonances of the vocal tract. In this paper, Mel-Frequency Cepstrum Coefficient (MFCC) is used to extract speakers features, and Gaussian Mixture Model (GMM) is established for the speakers, Maximum Likelihood Estimate (MLE) is use to estimate the model parameters of GMM. The aim is to identify the Cantonese speaker quickly and accurately in the noisy environment. Experimental results indicate that the method is very effective under noisy environments.