Speech Communication from an Information Theoretical Perspective

<p>Throughout the last century, models of human speech communication have been proposed by linguists, psychologists, and engineers. Advancements have been made, but a theory of human speech communication that is both comprehensive and quantitative is yet to emerge. This thesis hypothesises that a branch of mathematics known as information theory holds the answer to a more complete theory. Information theory has made fundamental contributions to wireless communications, computer science, statistical inference, cryptography, thermodynamics, and biology. There is no reason that information theory cannot be applied to human speech communication, but thus far, a relatively small effort has been made to do so. The goal of this research was to develop a quantitative model of speech communication that is consistent with our knowledge of linguistics and that is accurate enough to predict the intelligibility of speech signals. Specifically, this thesis focuses on the following research questions: 1) how does the acoustic information rate of speech compare to the lexical information rate of speech? 2) How can information theory be used to predict the intelligibility of speech-based communication systems? 3) How well do competing models of speech communication predict intelligibility? To answer the first research question, novel approaches for estimating the information rate of speech communication are proposed. Unlike existing approaches, the methods proposed in this thesis rely on having a chorus of speech signals where each signal in the chorus contains the same linguistic message, but is spoken by a different talker. The advantage of this approach is that variability inherent in the production of speech can be accounted for. The approach gives an estimate of about 180 b/s. This is three times larger than estimates based on lexical models, but it is an order of magnitude smaller than previous estimates that rely on acoustic signals. To answer the second research question, a novel instrumental intelligibility metric called speech intelligibility in bits (SIIB) and a variant called SIIBGauss are proposed. SIIB is an estimate of the amount of information shared between a talker and a listener in bits per second. Unlike existing intelligibility metrics that are based on information theory, SIIB accounts for talker variability and statistical dependencies between time-frequency units. Finally, to answer the third research question, a comprehensive evaluation of intrusive intelligibility metrics is provided. The results show that SIIB and SIIBGauss have state-of-the-art performance, that intelligibility metrics tend to perform poorly on data sets that were not used during their development, and show the advantage of reducing statistical dependencies between input features.</p>

Speech Communication from an Information Theoretical Perspective

<p>Throughout the last century, models of human speech communication have been proposed by linguists, psychologists, and engineers. Advancements have been made, but a theory of human speech communication that is both comprehensive and quantitative is yet to emerge. This thesis hypothesises that a branch of mathematics known as information theory holds the answer to a more complete theory. Information theory has made fundamental contributions to wireless communications, computer science, statistical inference, cryptography, thermodynamics, and biology. There is no reason that information theory cannot be applied to human speech communication, but thus far, a relatively small effort has been made to do so. The goal of this research was to develop a quantitative model of speech communication that is consistent with our knowledge of linguistics and that is accurate enough to predict the intelligibility of speech signals. Specifically, this thesis focuses on the following research questions: 1) how does the acoustic information rate of speech compare to the lexical information rate of speech? 2) How can information theory be used to predict the intelligibility of speech-based communication systems? 3) How well do competing models of speech communication predict intelligibility? To answer the first research question, novel approaches for estimating the information rate of speech communication are proposed. Unlike existing approaches, the methods proposed in this thesis rely on having a chorus of speech signals where each signal in the chorus contains the same linguistic message, but is spoken by a different talker. The advantage of this approach is that variability inherent in the production of speech can be accounted for. The approach gives an estimate of about 180 b/s. This is three times larger than estimates based on lexical models, but it is an order of magnitude smaller than previous estimates that rely on acoustic signals. To answer the second research question, a novel instrumental intelligibility metric called speech intelligibility in bits (SIIB) and a variant called SIIBGauss are proposed. SIIB is an estimate of the amount of information shared between a talker and a listener in bits per second. Unlike existing intelligibility metrics that are based on information theory, SIIB accounts for talker variability and statistical dependencies between time-frequency units. Finally, to answer the third research question, a comprehensive evaluation of intrusive intelligibility metrics is provided. The results show that SIIB and SIIBGauss have state-of-the-art performance, that intelligibility metrics tend to perform poorly on data sets that were not used during their development, and show the advantage of reducing statistical dependencies between input features.</p>

Extensions of Markov Modulated Poisson Processes and Their Applications to Deep Earthquakes

<p>The focus of this thesis is on the Markov modulated Poisson process (MMPP) and its extensions, aiming to propose appropriate statistical models for the occurrence patterns of main New Zealand deep earthquakes. Such an attempt might be beyond the scope of the MMPP and its extensions, however we hope its main patterns can be characterized by current models proposed in three parts of the thesis. The first part of the thesis is concerned with introductions and preliminaries of discrete time hidden Markov models (HMMs) and MMPP. The exibility in model formulation and openness in model framework of HMMs are reviewed in this part, suggesting also possible extensions of MMPP. The second part of the thesis is mainly about several extensions of MMPP. One extension of MMPP is by associating each occurrence of MMPP with a mark. Such an extension is potentially useful for spatial-temporal modelling or other point processes with marks. A special case of this type of extension is by allowing the multiple observations of MMPP synchronized together under the same Markov chain. This extension opens the possibility of modelling multiple point process observations with weak dependence. The third extension is motivated by the attempt to describe small scale temporal clustering existing in the deep earthquakes via treating the recognized aftershocks as marks which itself forms a finite point process. The rest of the second part focuses on some information theoretical aspects of MMPPs such as the entropy rate of the underlying Markov chain and observed point process respectively and their mutual information rate. A conjecture on the possible links between mutual information rate of MMPP and the Fisher information of the estimated parameters is suggested. The second part on extensions of MMPP is featured by the derivation of the likelihood and complete likelihood, parameter estimation via EM algorithm, state smoothing estimation and model evaluation through systematic applications of rescaling theory of multivariate point processes and marked point processes. The third part of the thesis includes the applications of these methods to the deep earthquakes in New Zealand. We first evaluate the data coverage, catalogue completeness and explore its descriptive characteristics and empirical properties such as epicentral distributions, depth distributions and magnitude distributions. Clustering behavior is studied via the second order moment analysis of point processes in the chapter 8. We also apply, the stress release models and the ETAS models which are usually used for shallow earthquakes, to the New Zealand deep earthquakes and provide tentative explanations of why they are not satisfactory for the deep earth-quakes. The chapter 9 is on the applications of MMPP and its extensions to the New Zealand deep earthquakes. Conclusions and future studies are presented in chapter 10.</p>

Extensions of Markov Modulated Poisson Processes and Their Applications to Deep Earthquakes

<p>The focus of this thesis is on the Markov modulated Poisson process (MMPP) and its extensions, aiming to propose appropriate statistical models for the occurrence patterns of main New Zealand deep earthquakes. Such an attempt might be beyond the scope of the MMPP and its extensions, however we hope its main patterns can be characterized by current models proposed in three parts of the thesis. The first part of the thesis is concerned with introductions and preliminaries of discrete time hidden Markov models (HMMs) and MMPP. The exibility in model formulation and openness in model framework of HMMs are reviewed in this part, suggesting also possible extensions of MMPP. The second part of the thesis is mainly about several extensions of MMPP. One extension of MMPP is by associating each occurrence of MMPP with a mark. Such an extension is potentially useful for spatial-temporal modelling or other point processes with marks. A special case of this type of extension is by allowing the multiple observations of MMPP synchronized together under the same Markov chain. This extension opens the possibility of modelling multiple point process observations with weak dependence. The third extension is motivated by the attempt to describe small scale temporal clustering existing in the deep earthquakes via treating the recognized aftershocks as marks which itself forms a finite point process. The rest of the second part focuses on some information theoretical aspects of MMPPs such as the entropy rate of the underlying Markov chain and observed point process respectively and their mutual information rate. A conjecture on the possible links between mutual information rate of MMPP and the Fisher information of the estimated parameters is suggested. The second part on extensions of MMPP is featured by the derivation of the likelihood and complete likelihood, parameter estimation via EM algorithm, state smoothing estimation and model evaluation through systematic applications of rescaling theory of multivariate point processes and marked point processes. The third part of the thesis includes the applications of these methods to the deep earthquakes in New Zealand. We first evaluate the data coverage, catalogue completeness and explore its descriptive characteristics and empirical properties such as epicentral distributions, depth distributions and magnitude distributions. Clustering behavior is studied via the second order moment analysis of point processes in the chapter 8. We also apply, the stress release models and the ETAS models which are usually used for shallow earthquakes, to the New Zealand deep earthquakes and provide tentative explanations of why they are not satisfactory for the deep earth-quakes. The chapter 9 is on the applications of MMPP and its extensions to the New Zealand deep earthquakes. Conclusions and future studies are presented in chapter 10.</p>

Causal Information Rate

Information processing is common in complex systems, and information geometric theory provides a useful tool to elucidate the characteristics of non-equilibrium processes, such as rare, extreme events, from the perspective of geometry. In particular, their time-evolutions can be viewed by the rate (information rate) at which new information is revealed (a new statistical state is accessed). In this paper, we extend this concept and develop a new information-geometric measure of causality by calculating the effect of one variable on the information rate of the other variable. We apply the proposed causal information rate to the Kramers equation and compare it with the entropy-based causality measure (information flow). Overall, the causal information rate is a sensitive method for identifying causal relations.

Optimal Wireless Information and Power Transfer Using Deep Q-Network

In this paper, a multiantenna wireless transmitter communicates with an information receiver while radiating RF energy to surrounding energy harvesters. The channel between the transceivers is known to the transmitter, but the channels between the transmitter and the energy harvesters are unknown to the transmitter. By designing its transmit covariance matrix, the transmitter fully charges the energy buffers of all energy harvesters in the shortest amount of time while maintaining the target information rate toward the receiver. At the beginning of each time slot, the transmitter determines the particular beam pattern to transmit with. Throughout the whole charging process, the transmitter does not estimate the energy harvesting channel vectors. Due to the high complexity of the system, we propose a novel deep Q-network algorithm to determine the optimal transmission strategy for complex systems. Simulation results show that deep Q-network is superior to the existing algorithms in terms of the time consumption to fulfill the wireless charging process.