A Novel Approach for Continuous Speech Tracking and Dynamic Time Warping. Adaptive Framing Based Continuous Speech Similarity Measure and Dynamic Time Warping using Kalman Filter and Dynamic State Model
KeywordSpeech tracking; Dynamic time warping; Kalman filter; Dynamic noise filtration; Adaptive framing; Keyword spotting; Template matching; Similarity measurement.
The University of Bradford theses are licenced under a Creative Commons Licence.
InstitutionUniversity of Bradford
DepartmentSchool of Electrical Engineering & Computer Science
MetadataShow full item record
AbstractDynamic speech properties such as time warping, silence removal and background noise interference are the most challenging issues in continuous speech signal matching. Among all of them, the time warped speech signal matching is of great interest and has been a tough challenge for the researchers. An adaptive framing based continuous speech tracking and similarity measurement approach is introduced in this work following a comprehensive research conducted in the diverse areas of speech processing. A dynamic state model is introduced based on system of linear motion equations which models the input (test) speech signal frame as a unidirectional moving object along the template speech signal. The most similar corresponding frame position in the template speech is estimated which is fused with a feature based similarity observation and the noise variances using a Kalman filter. The Kalman filter provides the final estimated frame position in the template speech at current time which is further used for prediction of a new frame size for the next step. In addition, a keyword spotting approach is proposed by introducing wavelet decomposition based dynamic noise filter and combination of beliefs. The Dempster’s theory of belief combination is deployed for the first time in relation to keyword spotting task. Performances for both; speech tracking and keyword spotting approaches are evaluated using the statistical metrics and gold standards for the binary classification. Experimental results proved the superiority of the proposed approaches over the existing methods.
NotesThe appendices files are not available online.
Showing items related by title, author, creator and subject.
Directing the Assembly of Multicomponent Organic Crystals. Synthesis, characterisation and structural analysis of multicomponent organic systems formed from dynamic processes.Scowen, Ian J.; Munshi, Tasnim; Alomar, Taghrid S. (University of BradfordSchool of Life Sciences, 2015-04-22)Directed assembly of molecular solids continues to attract widespread interest with its fundamental application in a wide range of commercial settings where control of the crystalline state of materials corresponds with product performance. These arenas include pharmaceuticals, personal care formulations, foods, paints and pigments and explosives. In recent times, the assembly of multicomponent organic systems has achieved considerable impetus with the widespread interest in co-crystal systems. However, cogent assembly (or engineering) of multicomponent materials is still in its infancy. Considerable advances in crystal design have been made through consideration of intermolecular ‘synthons’ – identifiable motifs utilising hydrogen bonds – but the translation of other molecular information (conformation, chirality, etc.) into solid state properties (e.g. long-range (translational) symmetry, crystal chirality) remains poorly understood. In this study, we have attempted to evaluate the influence of a chiral centre adjacent to molecular synthons to identify potential translation of information into the solid form. We have compared the co-crystallisation of nicotinamide with both the racemic mixture of malic acid against that with an enantiomerically pure form of the acid (L-malic acid). As well as DL-phenyllactic acid and L-phenyllactic acid. iii It is apparent that recognition between enantiomeric molecular forms play a significant role in the assembly of these systems. This mechanism can be considered independently from the H-bonding networks supporting the hetero-molecular interactions (e.g. acid-amide recognition). Discrimination and control of such interactions may play a role in transmitting chiral molecular information into solid state multi-component assemblies. In order to develop an understanding of co-crystal formation in chiral and achiral forms with intermolecular interactions, the CSD and crystal structures were obtained to do the analysis of how co-crystals pack. This study has also investigated the use of boronic acids. The aim of this study was to investigate the modification of the hydrogen bonding environment within the hydrogen bonded multi-component systems of boroxines. The study also attempted to determine how the starting materials drive the systems between the boronic acid co-crystal and the boroxine adduct.
New Dynamic Rotamer Libraries: Data-Driven Analysis of Side-Chain Conformational PropensitiesTowse, Clare-Louise; Rysavy, S.J.; Vulovic, I.M.; Daggett, V. (2016-01-05)Most rotamer libraries are generated from subsets of the PDB and do not fully represent the conformational scope of protein side chains. Previous attempts to rectify this sparse coverage of conformational space have involved application of weighting and smoothing functions. We resolve these limitations by using physics-based molecular dynamics simulations to determine more accurate frequencies of rotameric states. This work forms part of our Dynameomics initiative and uses a set of 807 proteins selected to represent 97% of known autonomous protein folds, thereby eliminating the bias toward common topologies found within the PDB. Our Dynameomics derived rotamer libraries encompass 4.8 × 10(9) rotamers, sampled from at least 51,000 occurrences of each of 93,642 residues. Here, we provide a backbone-dependent rotamer library, based on secondary structure ϕ/ψ regions, and an update to our 2011 backbone-independent library that addresses the doubling of our dataset since its original publication.
Dynamic modeling of a transient engine test cell for cold engine testing applicationsSouflas, Ioannis; Pezouvanis, Antonios; Mason, Byron A.; Ebrahimi, Kambiz M. (2014)The increasing complexity in the development and manufacturing process of internal combustion engines leads to a higher demand for more effective testing and monitoring methods. Cold engine testing becomes progressively the main End-of-Line test which is used nowadays from automotive engine manufacturers with the purpose of determining the integrity of engine assembly. The present work is focused on the development of a detailed physics-based, lumped-parameter, dynamic model of a single cylinder internal combustion engine coupled with an alternating current transient dynamometer for cold engine testing applications. The overall transient engine test cell model is described based on a two-inertia system model consisting of the engine, the dynamometer and the coupling shaft. The internal combustion engine is modelled based on First Law of Thermodynamics and Second Newton’s Law for rotational bodies. The transient dynamometer is actually an alternating current three-phase induction motor which is modelled according to direct-quadrature axis approach, and its drive unit which is responsible for controlling the speed of the motor using indirect field orientation scheme. The engine and dynamometer are connected through a coupling shaft which is modelled as a compliant member with damping. The model is validated against experimental measurements such as engine cylinder pressure, engine excitation torque and alternating currents of the induction motor. All of the experimental measurements were recorded from an identical single cylinder transient engine test cell using a highly advanced instrumentation system. The described model serves as an ideal platform for developing innovative model-based fault detection and diagnosis techniques for cold engine testing applications. In conclusion, this is presented successfully for two simulated fault cases, a process fault and a sensor fault, proving the functionality and usefulness of the model.