Global ETD Search

Search theses and dissertations gathered from participating repositories worldwide. Every result links back to the library that holds it. No account is needed.

Results

Showing 1 to 20 of 104 for “"Acoustic models"”.

  1. Deep Neural Network Acoustic Models for ASR

    … unchanged for many years. It used Hidden Markov Models (HMMs) to model the sequential structure of speech signals, with each HMM state using a mixture of diagonal covariance Gaussians (GMM) to model a spectral representation of the sound wave. This thesis describes new acoustic models based on …

    toronto-retro Repository record for Deep Neural Network Acoustic Models for ASR (opens in a new tab)

  2. Implementation of DNN-HMM Acoustic Models for Phoneme Recognition

    Gaussian Mixture Model-Hidden Markov Models (GMM-HMMs) are the state-of-the-art for acoustic modeling in speech recognition. HMMs are used to model the sequential structure and the temporal variability in speech signals. However, GMMs are used to model the local spectral variability in the sound …

    waterloo Repository record for Implementation of DNN-HMM Acoustic Models for Phoneme Recognition (opens in a new tab)

  3. Acoustic models of consonant recognition in cochlear implant users

    … made by normal-hearing subjects listening to acoustic models of implant processing, in various conditions. Two initial experiments were undertaken to develop and refine an acoustic model of the Nucleus 24 cochlear implant. Findings from these two experiments informed the design of the main …

    soton Download PDF for Acoustic models of consonant recognition in cochlear implant users (opens in a new tab)

  4. Hadamard Well-Posedness for Two Nonlinear Structure Acoustic Models

    … well-posedness of two nonlinear structure acoustic models, each consisting of a semilinear wave equation defined on a smooth bounded domain $\Omega\subset\mathbb{R}^3$ strongly coupled with a Berger plate equation acting only on a flat portion of the boundary of $\Omega$. In each case, the …

    nebraska Repository record for Hadamard Well-Posedness for Two Nonlinear Structure Acoustic Models (opens in a new tab)

  5. Statistical optimization of acoustic models for large vocabulary speech recognition

    This dissertation investigates optimization of acoustic models in speech recognition. Two new optimization methods are proposed for phonetic decision tree (PDT) search and Hidden Markov modeling (HMM)-- the knowledge-based adaptive PDT algorithm and the HMM gradient boosting algorithm. …

    missouri Repository record for Statistical optimization of acoustic models for large vocabulary speech recognition (opens in a new tab)

  6. Acoustic Models for the Analysis and Synthesis of the Singing Voice

    … have been developed for normal speech, few models and techniques have been focused on the singing voice. The central theme of this research is the development of models aimed at the characterization and synthesis of the singing voice. First, a spectral model is presented in which asymmetric …

    gatech Repository record for Acoustic Models for the Analysis and Synthesis of the Singing Voice (opens in a new tab)

  7. A comparative study of acoustic models in a range-independent shallow water environment

    … data set was analyzed and compared to various acoustic propagation models approved or under consideration for inclusion in the Navy's Ocean Atmospheric Master Library (OAML). The following models were considered in the analysis: PE, utilizing the OAML approved LFBL database, FEPE and SNAP. FEPE …

    nps Repository record for A comparative study of acoustic models in a range-independent shallow water environment (opens in a new tab)

  8. Comparison of Finite-amplitude Acoustic Models: Single Equation of Motion for Potential vs System-level Approaches

    This study evaluates two finite-amplitude acoustic models, by solving the associated equations of motion (in potential form), and then compares the results with recent studies using hyperbolic system-level approaches. An implicit numerical method is employed to solve the second-order equations in …

    uno Repository record for Comparison of Finite-amplitude Acoustic Models: Single Equation of Motion for Potential vs System-level Approaches (opens in a new tab)

  9. Kalbos signalų akustinių modelių, skirtų kalbėtojo atpažinimui, tyrimas /

    … transcription. The research of speech signal acoustic models for speaker recognition been described in this paper. The aim of this thesis – to investigate acoustic speech signal models suitable for speaker recognition. In the analytical practical part voice records were investigate, MFCC …

    vilnius Repository record for Kalbos signalų akustinių modelių, skirtų kalbėtojo atpažinimui, tyrimas / (opens in a new tab)

  10. Automatinio šnekos atpažinimo metodų tyrimas ir taikymai balso įrašams stenografuoti /

    … In the flow of this work, Hidden Markov Models (acoustic models), language model and dictionary were created. These models was integrated into voice records stenography application. Based on testing results of acoustic models an accuracy of 94% was achieved.

    vilnius Repository record for Automatinio šnekos atpažinimo metodų tyrimas ir taikymai balso įrašams stenografuoti / (opens in a new tab)

  11. A comparison of passive propagation loss predictions as generated by the Tactical Environmental Support System (TESS v.2.2A) and ASW Tactical Decision Aid (ASWTDA v.2.1.2.1).

    … of different systems to predict underwater acoustic transmission loss for operational forces. Historically, these systems have used different acoustic models and supporting databases, resulting in significantly different predictions. Major efforts to bring all acoustic models and databases …

    nps Repository record for A comparison of passive propagation loss predictions as generated by the Tactical Environmental Support System (TESS v.2.2A) and ASW Tactical Decision Aid (ASWTDA v.2.1.2.1). (opens in a new tab)

  12. Towards Efficient and Robust Automatic Speech Recognition: Decoding Techniques and Discriminative Training

    … of speech recognition, decoding and training of acoustic models, to improve speech recognition performance in different conditions. A major part of this thesis studies discriminative training of acoustic models. The emphasis is on the most popular algorithm for discriminative model estimation, …

    aalto Repository record for Towards Efficient and Robust Automatic Speech Recognition: Decoding Techniques and Discriminative Training (opens in a new tab)

  13. Evaluation of Acoustic Frequency Methods Coupled to Blade Element Momentum Theory for the Prediction of Propeller Noise

    … accuracy of several computationally-inexpensive acoustic frequency methods is evaluated across a range of propeller geometries and operational conditions. The acoustic models considered do not require chord-wise aerodynamic data and therefore do not need to be coupled to a panel or grid-based …

    carleton Repository record for Evaluation of Acoustic Frequency Methods Coupled to Blade Element Momentum Theory for the Prediction of Propeller Noise (opens in a new tab)

  14. Integral equation methods for the simulation of viscoelastic ultrasound vibro-acoustography.

    … had already been obtained for simplified (purely acoustic) models of UVA (where the human tissue is modeled as fluid-like material), there were certainly possibilities for further improvements. In this connection, the work in this thesis focuses on the derivation of more realistic viscoelastic UVA …

    umn Repository record for Integral equation methods for the simulation of viscoelastic ultrasound vibro-acoustography. (opens in a new tab)

  15. Multi-Task and Transfer Learning in Low-Resource Speech Recognition

    This thesis investigates methods for Acoustic Modeling in Automatic Speech Recog- nition, assuming limited access to training data in the target domain. The Acoustic Models of interest are Deep Neural Network Acoustic Models (in both the Hybrid and End-to-End approaches), and the target domains in …

    arizona-thes Repository record for Multi-Task and Transfer Learning in Low-Resource Speech Recognition (opens in a new tab)

  16. Ensemble acoustic modeling in automatic speech recognition

    … using data sampling to construct an Ensemble of Acoustic Models (EAM) for speech recognition are proposed. A straightforward method of data sampling is Cross Validation (CV) data partition. In the direction of improving inter-model diversity within an EAM for speaker independent speech …

    missouri Repository record for Ensemble acoustic modeling in automatic speech recognition (opens in a new tab)

  17. Comparative analysis of the SAFARI model with NATIVE 1 data in the VLF regime

    Few underwater acoustic models have been validated in the Very Low Frequency (VLF) regime. The purpose here is to explore one model's abilities and compare it with ocean data once it is ascertained the model has converged to its final solution. A comparative analysis of data recorded in NATIVE 1 …

    nps Repository record for Comparative analysis of the SAFARI model with NATIVE 1 data in the VLF regime (opens in a new tab)

  18. Data sufficiency analysis for automatic speech recognition

    … area. Specifically, current ASR systems utilise acoustic models to represent acoustic variability, and effective ASR corpus design aims to optimise the amount of relevant variation within training data while minimising the size of the corpus. Therefore an investigation of the effect that …

    nwu-za Repository record for Data sufficiency analysis for automatic speech recognition (opens in a new tab)

  19. Control of broadband acoustic radiation from structures using a piezoelectric double-amplifier active-skin

    … is demonstrated for the reduction of broadband acoustic radiation from a vibrating structure. A simplified representation of the active-skin, employing acoustic monopoles as secondary sources, is explored as a precursor to the more complicated analyses of the device. Many design issues are …

    vt Repository record for Control of broadband acoustic radiation from structures using a piezoelectric double-amplifier active-skin (opens in a new tab)

  20. Non-linear Cross-lingual Mappings for Low-resource Speech Recognition

    … recognition systems and their relation with acoustic-phonetic similarities across the languages. The objective of this research is to estimate cross-lingual acoustic-phonetic similarities and their impact on multilingual speech recognition systems. To that end, a novel data-driven approach is …

    whiterose Download PDF for Non-linear Cross-lingual Mappings for Low-resource Speech Recognition (opens in a new tab)

Page 1 of 6