Narrow Results

Most of the Indian-language Text-To-Speech (TTS) synthesis systems designed till date are based upon the concatenation of acoustic units. The prime challenge is the selection of proper units and their elegant concatenation. Due to the precincts of current automated techniques based on Hidden Markov Model (HMM) and Dynamic Time Warping (DTW), manual verification and labeling are often essential. Automatic placement of phoneme boundaries in a speech waveform using explicit statistical model for phoneme boundary is proposed in this paper. We are projecting the Harmonic plus Noise Model (HNM) in the first step and refine the boundary placement by searching for the best match in a region near the estimated boundary with predefined boundary model Technique like ESNOLA. This technique is applied for effective concatenation, which results in smooth output. Studies show that HNM is capable of synthesizing all vowels and diphones with good quality. This can remarkably reduce the size of the database. Further the pitch synchronous analysis and the Glottal Closure Instants (GCI) are accurately calculated. The quality of the synthesized speech improves if these units are obtained from the glottal signal rather than from processing the signal. The database has to be developed for VCV for all Indian languages as we have done for Oriya, one of the official languages of the Republic of India for our case study.

Most parameter-based online signature verification methods achieve correspondence between the points of two signatures by minimizing the accumulation of their local feature distances. The matching based on minimizing the local feature distances alone is not adequate since the point contains not only local features but the distribution of the remaining points relative to it. One useful way to get correspondences between points on two shapes and measure the similarity of the two shapes is to use the shape context, since this descriptor can be used to describe the distributive relationship between a reference point and the remaining points on a shape. In this paper, we introduce a shape context descriptor for describing an online signature point which contains both 2D spatial information and a time stamp. A common algorithm, dynamic time warp (DTW), is used for the elastic matching between two signatures. When combining shape contexts and local features, we achieve better results than when using only the local features. We evaluate the proposed method on a signature database from the First International Signature Verification Competition (SVC2004). Experimental results demonstrate that the shape context is a good feature and has available complementarity for describing the signature point. The best result by combining the shape contexts and the local features yields an Equal Error Rate (EER) of 6.77% for five references.

In this paper, we propose a new scheme for Devanagari natural handwritten character recognition. It is primarily based on spatial similarity-based stroke clustering. A feature of a stroke consists of a string of pen-tip positions and directions at every pen-tip position along the trajectory. It uses the dynamic time warping algorithm to align handwritten strokes with stored stroke templates and determine their similarity. Experiments are carried out with the help of 25 native writers and a recognition rate of approximately 95% is achieved. Our recognizer is robust to a large range of writing style and handles variation in the number of strokes, their order, shapes and sizes and similarities among classes.

Although the task-specific robot treadmill therapies can improve gait performance of children with cerebral palsy, not all the institutions have the possibility to access to the sensor robot information to obtain quantitative measures related actual progress of the children. The aim of this study is to propose a computer vision system to investigate the effort of children with cerebral palsy during robot treadmill therapy. Ten participants diagnosed with cerebral palsy underwent single robot therapy sessions consisting of a motivated period (active: kicking a ball), followed by an unmotivated period (passive: simple walking). The outcome measure was the ankle trajectory recorded by a video camera. Indexes based on the ankle trajectory between active and passive therapy were computed: (a) the gait analysis based on the difference on a sample-by-sample manner, and b) the step cycles with dynamic time warping (DTW). Inference statistical analysis shows that the robot trajectories for passive and active therapies with the gait analysis presents statistical differences (p-$\mathrm{\text{value}}<0.01))$. Likewise, DTW showed significant differences in robot trajectories related to passive and active therapy (p-$\mathrm{\text{value}}<0.01))$. We found that the statistical differences are independent of the severity of central gait impairment. This suggests that information about patient involvement could be given to the therapist and could be used to assess patient participation. The method showed the potential to be used as a tool for the assessment the improvement progress.

To better understand the dynamics of the underlying processes in cells, it is necessary to take measurements over a time course. Modern high-throughput technologies are often used for this purpose to measure the behavior of cell products like metabolites, peptides, proteins, $\mu$RNA or mRNA at different points in time. Compared to classical time series, the number of time points is usually very limited and the measurements are taken at irregular time intervals. The main reasons for this are the costs of the experiments and the fact that the dynamic behavior usually shows a strong reaction and fast changes shortly after a stimulus and then slowly converges to a certain stable state. Another reason might simply be missing values. It is common to repeat the experiments and to have replicates in order to carry out a more reliable analysis. The ideal assumptions that the initial stimulus really started exactly at the same time for all replicates and that the replicates are perfectly synchronized are seldom satisfied. Therefore, there is a need to first adjust or align the time-resolved data before further analysis is carried out. Dynamic time warping (DTW) is considered as one of the common alignment techniques for time series data with equidistant time points. In this paper, we modified the DTW algorithm so that it can align sequences with measurements at different, non-equidistant time points with large gaps in between. This type of data is usually known as time-resolved data characterized by irregular time intervals between measurements as well as non-identical time points for different replicates. This new algorithm can be easily used to align time-resolved data from high-throughput experiments and to come across existing problems such as time scarcity and existing noise in the measurements. We propose a modified method of DTW to adapt requirements imposed by time-resolved data by use of monotone cubic interpolation splines. Our presented approach provides a nonlinear alignment of two sequences that neither need to have equi-distant time points nor measurements at identical time points. The proposed method is evaluated with artificial as well as real data. The software is available as an R package tra (Time-Resolved data Alignment) which is freely available at: http://public.ostfalia.de/klawonn/tra.zip.

Cough recognition provides important clinical information for the treatment of many respiratory diseases, but the assessment of cough frequency over a long period of time remains unsatisfied for either clinical or research purpose. In this paper, according to the advantage of dynamic time warping (DTW) and the characteristic of cough recognition, an attempt is made to adapt DTW as the recognition algorithm for cough recognition. The process of cough recognition based on mel frequency cepstral coefficients (MFCC) and DTW is introduced. Experiment results of testing samples from 3 subjects show that acceptable performances of cough recognition are obtained by DTW with a small training set.