In this work, a dynamic speech parameterization based on the continuous multiresolution divergence is used to modify a text-independent phone segmentation algorithm. This encoding is employed as input and also replaces an stage of the segmentation procedure responsible for the estimation of the intensity of changes in signal features. The segmentation performance of this representation has been compared with the original algorithm using as input a classical Melbank parameterization and speech representation based on the continuous multiresolution divergence. The results indicate that the modification here proposed increases the ability of the algorithm to perform the segmentation task. This suggests that continuous multiresolution divergence provides valuable information related to acoustic features that take into account phoneme transitions. Moreover, this parameterization gives enough information for its direct use without further processing.
Download full-text PDF |
Source |
|---|---|
| http://dx.doi.org/10.1109/IEMBS.2010.5628010 | DOI Listing |
Publication Analysis
Top Keywords
Similar Publications
CoCoNest: A continuous structural connectivity-based nested family of parcellations of the human cerebral cortex.
Netw Neurosci
December 2024
Statistics and Operations Research, University of North Carolina at Chapel Hill, Chapel Hill, NC, USA.
Despite the widespread exploration and availability of parcellations for the functional connectome, parcellations designed for the structural connectome are comparatively limited. Current research suggests that there may be no single "correct" parcellation and that the human brain is intrinsically a multiresolution entity. In this work, we propose the Continuous Structural Connectivitity-based, Nested (CoCoNest) family of parcellations-a fully data-driven, multiresolution family of parcellations derived from structural connectome data.
View Article and Find Full Text PDFContinuous Generative Neural Networks: A Wavelet-Based Architecture in Function Spaces.
Numer Funct Anal Optim
November 2024
MaLGa Center, Department of Mathematics, University of Genoa, Genova , Italy.
In this work, we present and study Continuous Generative Neural Networks (CGNNs), namely, generative models in the continuous setting: the output of a CGNN belongs to an infinite-dimensional function space. The architecture is inspired by DCGAN, with one fully connected layer, several convolutional layers and nonlinear activation functions. In the continuous setting, the dimensions of the spaces of each layer are replaced by the scales of a multiresolution analysis of a compactly supported wavelet.
View Article and Find Full Text PDFApplication of big data and artificial intelligence in visual communication art design.
PeerJ Comput Sci
November 2024
Shi Jia Zhuang University of Applied Technology, Shijiazhuang, Hebei, China.
In the era of continuous development of computer technology, the application of artificial intelligence (AI) and big data is becoming more and more extensive. With the help of powerful computer and network technology, the art of visual communication (VISCOM) has ushered in a new chapter of digitalization and intelligence. How vision can better perform interdisciplinary and interdisciplinary artistic expression between art and technology and how to use more novel technology, richer forms, and more appropriate ways to express art has become a new problem in visual art creation.
View Article and Find Full Text PDFMV-GNN: Generation of continuous geometric representations of mitral valve motion from 3D+t echocardiography.
Comput Biol Med
November 2024
Institute for Imaging Science and Computational Modelling in Cardiovascular Medicine, Charité - Universitätsmedizin Berlin, 13353 Berlin, Germany; Deutsches Herzzentrum der Charité, 13353 Berlin, Germany; Fraunhofer MEVIS, Max-von-Laue-Str. 2, 28359 Bremen, Germany; DZHK (German Centre for Cardiovascular Research), Berlin, Germany.
We present a geometric deep-learning method for reconstructing a temporally continuous mitral valve surface mesh from 3D transesophageal echocardiography sequences. Our approach features a supervised end-to-end deep learning architecture that combines a convolutional neural network-based voxel encoder and decoder with a graph neural network-based multi-resolution mesh decoder, all trained on sparse landmark annotations. Key elements of our methodology include a tube-shaped prototype mesh with labeled vertices, a specialized loss function to preserve the known inlet and outlet, and a rigid alignment system for anatomical landmarks.
View Article and Find Full Text PDFAn Information Bottleneck Approach for Markov Model Construction.
ArXiv
June 2024
Department of Chemistry and Biochemistry and Institute for Physical Science and Technology, University of Maryland, College Park, MD 20742, United States.
Markov state models (MSMs) have proven valuable in studying dynamics of protein conformational changes via statistical analysis of molecular dynamics (MD) simulations. In MSMs, the complex configuration space is coarse-grained into conformational states, with dynamics modeled by a series of Markovian transitions among these states at discrete lag times. Constructing the Markovian model at a specific lag time necessitates defining states that circumvent significant internal energy barriers, enabling internal dynamics relaxation within the lag time.
View Article and Find Full Text PDFWant AI Summaries of new PubMed Abstracts delivered to your In-box?
Enter search terms and have AI summaries delivered each week - change queries or unsubscribe any time!