Volume 22, Issue 3 - Special Issue: International Conference on Communication, VLSI and Signal Processing 2013, SIT Tumkur, India

The classification of unvoiced stops in consonant–vowel (CV) syllables, segmented from continuous speech, is investigated by features related to speech production. As burst and vocalic transitions contribute to identification of stops in the CV context, features are computed from both regions. Although formants are the truly discriminating articulatory features, their estimation from the speech signal is a challenge especially in unvoiced regions like the release burst of stops. This may be compensated partially by sub-band energy-based features. In this work, formant features from the vocalic region are combined with features from the burst region comprising sub-band energies, as well as features from a formant tracking method developed for unvoiced regions. The overall combination of features at the classifier level obtains an accuracy of 84.4%, which is significantly better than that obtained with solely sub-band features on unvoiced stops in CV syllables of TIMIT.

Rapid handwriting, popularly known as shorthand, involves writing symbols and abbreviations in lieu of common words or phrases. This method increases the speed of transcription and is primarily used to record oral dictation. Someone skilled in shorthand will be able to write as fast as the dictation occurs, and these patterns are later transliterated into actual, natural language words. A new kind of rapid handwriting scheme is proposed, called the Pattern-Based Shorthand. A word on a keyboard involves pressing a unique sequence of keys in a particular order. This sequence forms a pattern that defines the word. Such a pattern forms the shorthand for that word. Speech recognition involves identifying, by a machine, the words spoken by a speaker. These spoken words form speech input signals to a computer that is equipped to correctly recognize the words and do further action, such as convert it to text. From this text input, unique shorthand patterns are generated by the system. The system employs machine learning to improve its performance with experience, by creating a dictionary of mappings from word to patterns in such a way that the access to existing patterns is faster with progression. This forms a new knowledge representation schema that reduces the redundancy in the storage of words and the length of information content. In conclusion, the speech is converted into textual form and then reconstructed into Pattern-Based Shorthand.

In this work, the significance of combining the evidence from multitaper mel-frequency cepstral coefficients (MFCC), linear prediction residual (LPR), and linear prediction residual phase (LPRP) features for multilingual speaker identification with the constraint of limited data condition is demonstrated. The LPR is derived from linear prediction analysis, and LPRP is obtained by dividing the LPR using its Hilbert envelope. The sine-weighted cepstrum estimators (SWCE) with six tapers are considered for multitaper MFCC feature extraction. The Gaussian mixture model–universal background model is used for modeling each speaker for different evidence. The evidence is then combined at scoring level to improve the performance. The monolingual, crosslingual, and multilingual speaker identification studies were conducted using 30 randomly selected speakers from the IITG multivariability speaker recognition database. The experimental results show that the combined evidence improves the performance by nearly 8–10% compared with individual evidence.

We propose an approach for texture segmentation based on weak supervised learning. The weak supervision implies that the user marks only a single small patch for each class in the input image. These patches are used for training. We employ the method of graph cuts for the segmentation task. Our work demonstrates that even under such weak training, texture segmentation can be achieved efficiently and with good accuracy via graph cuts. Moreover, our approach uses a simpler feature representation than that in similar contemporary segmentation approaches. We also provide a brief discussion indicating the good performance of our approach. We validate our method on various standard texture mosaics and also on segmentation of natural images with large texture variations.

In this article, a new adaptive data-driven strategy for voice activity detection (VAD) using empirical mode decomposition (EMD) is proposed. Speech data are decomposed using an a posteriori, adaptive, data-driven EMD in the time domain to yield a set of physically meaningful intrinsic mode functions (IMFs). Each IMF preserves the nonlinear and nonstationary property of the speech utterance. Among a set of IMFs, the IMF that contains source information dominantly called characteristic IMF (CIMF) can be identified and extracted by designing a zero-frequency filter-assisted peaking resonator. The detected CIMF is used to compute energy using short-term processing. Choosing proper threshold, voiced regions in speech utterances are detected using frame energy. The proposed framework has been studied on both clean speech utterance and noisy speech utterance (0-dB white noise). The proposed method is used for voice activity detection (VAD) in the presence of white noise and shows encouraging result in the presence of white noise up to 0 dB.

This article describes the aircraft image recognition system implemented using the phase correlation technique in Matlab environment. The phase correlation is computed by using the normalized cross-power spectrum between the database and the template test image. The main objective of this article is to develop methods for static analysis of aircraft images. An unknown fighter aircraft is recognized by comparing its static image with those from a database of images of aircraft. This work is a research initiative involving the use of image processing techniques to detect three-dimensional (3D) aircraft object based on their 2D images, providing feedback information for strategic purposes. The phase correlation technique is found to give a better recognition result for the set of database and test images considered, compared with the invariant moments. The phase correlation method is also used in other areas such as image registration. Aircraft images used include those from AeroIndia 2011 held at Bangalore, India.

In this article, the covariance of wavelet transform coefficients is used to obtain a texture inpainted image. The covariance is obtained by the maximum likelihood estimate. The integration of wavelet decomposition and maximum likelihood estimate of a group of pixels (texture) captures the best-fitting texture used to fill in the inpainting area. The image is decomposed into four wavelet coefficient images by using wavelet transform. These wavelet coefficient images are divided into small square patches. The damaged region in each coefficient image is filled by searching a similar square patch around the damaged region in that particular wavelet coefficient image by using covariance.

This article deals with detection and accommodation of sensor faults in longitudinal dynamics of an F8 aircraft model. Both the detection of the fault and reconfiguration of the failed sensor are done with the help of neural network-based models. Detection of a sensor fault is done with the help of knowledge-based neural network fault detection (KBNNFD). Apart from KBNNFD, another neural network model is developed in this article for the reconfiguration of the failed sensor. A model-based approach of the neural network (MBNN) is developed, which uses the radial basis function of the neural network. MBNN successfully does the task of providing analytical redundancy for the aircraft sensor. In this work, both detection and reconfiguration of a fault is done using neural networks. Hence, the control system becomes robust for handling sensor failures near steady state and reconfiguration is also faster. A generalized regression neural network (GRNN), which is a type of radial basis network, is used for MBNN, which gives very efficient results for function approximation. An F8 aircraft model and C-Star controller, which improves its handling quality, are used for validation of the method involved. Models of F8 aircraft, C-Star controller, KBNNFD, and MBNN were developed using MATLAB/Simulink. Successful implementation and simulation results are shown and discussed using Simulink.

A large number of articles have been devoted to the application of “texture synthesis” for large regions and “inpainting” algorithms for small cracks in an image. A new approach that allows the simultaneous filling in of different structures and textures is discussed in this present study. The combination of structure inpainting and patch-based texture synthesis carried out (termed as “patch-based inpainting”) for filling and updating the target region shows additional advantages over earlier approaches. The algorithm discussed here uses the patch-based inpainting with isophote-driven patch-based texture synthesis at the core. In this algorithm, once the user selects the regions to be restored, the algorithm automatically searches and fills in these regions with the best matching information surrounding them. We have assigned high priorities to the pixels on the boundary and the structure by computing data terms D ( p ), and the texture and corners are prioritized by computing the confidence C ( p ) of the pixel. We also regularized and weighted the confidence of the pixels, RC ( p ), to achieve a balance of the two. The patch search area near the pixel patch to be filled is bounded for algorithm speed improvement. Patch-based filling significantly improve execution speed compared with pixel-based filling. Filling in is done in such a way that the structure information arriving at the region boundaries is propagated inside. A number of examples on real and synthetic images are used to demonstrate the effectiveness of the algorithm. Robustness with respect to the shape of the selected target region is also demonstrated.

In this article, epoch-based dynamic features such as sequence of epoch interval values and epoch strength values are explored to classify infant cries. Epoch is the instant of significant excitation of the vocal tract system during the production of speech. For voiced speech, the most significant excitation takes place around the instant of glottal closure. The different types of infant cries considered in this work are hunger, pain, and wet diaper. In this work, epoch strength and epoch interval features are used to represent infant cry-specific information from the acoustic signal. In this study, the proposed features such as epoch interval and epoch strength values are determined using zero-frequency filter-based method. Gaussian mixture models (GMMs) are used to classify the above-mentioned cries from the features proposed in this work. GMMs are developed separately for each of the cries using the proposed features. The infant cry database collected under a telemedicine project at the Indian Institute of Technology Kharagpur has been used for this study. In the first step, infant cry recognition accuracy is investigated separately using epoch interval and epoch strength features. To enhance recognition performance, GMMs developed using various features are combined through score level fusion techniques. The recognition performance using a combination of evidence is found to be superior over individual systems.