Unsupervised speaker adaptation for telephone call transcription

The use of the PC and Internet for placing telephone calls will present new opportunities to capture vast amounts of un-transcribed speech for a particular speaker. This paper investigates how to best exploit this data for speaker-dependent speech recognition. Supervised and unsupervised experiments in acoustic model and language model adaptation are presented. Using one hour of automatically transcribed speech per speaker with a word error rate of 36.0%, unsupervised adaptation resulted in an absolute gain of 6.3%, equivalent to 70% of the gain from the supervised case, with additional adaptation data likely to yield further improvements. LM adaptation experiments suggested that although there seems to be a small degree of speaker idiolect, adaptation to the speaker alone, without considering the topic of the conversation, is in itself unlikely to improve transcription accuracy.

Vibration of ship structures and its control

Noise and vibration in complex ship structures are becoming a prominent issue for ship building industry and ship companies due to the constant demand of building faster ships of lighter weight, and the stringent noise and libration regulation of the industry. In order to retain the full benefit of building faster ships without compromising too much on ride comfort and safety, noise and vibration control needs to be implemented. Due to the complexity of ship structures, the coupling of different wave types and multiple wave propagation paths, active control of global hull modes is difficult to implement and very expensive. Traditional passive control such as adding damping materials is only effective in the high frequency range. However, most severe damage to ship structures is caused by large structural deformation of hull structures and high dynamic stress concentration at low frequencies. The most discomfort and fatigue of passengers and the crew onboard ships is also due to the low frequency noise and vibration. Innovative approaches are therefore, required to attenuate the noise and vibration at low frequencies. This book was developed from several specialized research topics on vibration and vibration control of ship structures, mostly from the author's own PhD work at the University of Western Australia. The book aims to provide a better understanding of vibration characteristics of ribbed plate structures, plate/plate coupled structures and the mechanism governing wave propagation and attenuation in periodic and irregular ribbed structures as well as in complex ship structures. The book is designed to be a reference book for ship builders, vibro-acoustic engineers and researchers. The author also hopes that the book can stimulate more exciting future work in this area of research. It is the author's humble desire that the book can be some use for those who purchase it. This book is divided into eight chapters. Each chapter focuses on providing solution to address a particular issue on vibration problems of ship structures. A brief summary of each chapter is given in the general introduction. All chapters are inter-dependent to each other to form an integration volume on the subject of vibration and vibration control of ship structures and alike. I am in debt to many people in completing this work. In particular, I would like to thank Professor J. Pan, Dr N.H. Farag, Dr K. Sum and many others from the University of Western Australia for useful advices and helps during my times at the University and beyond. I would also like to thank my wife, Miaoling Wang, my children, Anita, Sophia and Angela Lin, for their sacrifice and continuing supports to make this work possible. Financial supports from Australian Research Council, Australian Defense Science and Technology Organization and Strategic Marine Pty Ltd at Western Australia for this work is gratefully acknowledged.

Acoustic keyword spotting in speech with applications to data mining

Keyword Spotting is the task of detecting keywords of interest within continu- ous speech. The applications of this technology range from call centre dialogue systems to covert speech surveillance devices. Keyword spotting is particularly well suited to data mining tasks such as real-time keyword monitoring and unre- stricted vocabulary audio document indexing. However, to date, many keyword spotting approaches have su®ered from poor detection rates, high false alarm rates, or slow execution times, thus reducing their commercial viability. This work investigates the application of keyword spotting to data mining tasks. The thesis makes a number of major contributions to the ¯eld of keyword spotting. The ¯rst major contribution is the development of a novel keyword veri¯cation method named Cohort Word Veri¯cation. This method combines high level lin- guistic information with cohort-based veri¯cation techniques to obtain dramatic improvements in veri¯cation performance, in particular for the problematic short duration target word class. The second major contribution is the development of a novel audio document indexing technique named Dynamic Match Lattice Spotting. This technique aug- ments lattice-based audio indexing principles with dynamic sequence matching techniques to provide robustness to erroneous lattice realisations. The resulting algorithm obtains signi¯cant improvement in detection rate over lattice-based audio document indexing while still maintaining extremely fast search speeds. The third major contribution is the study of multiple veri¯er fusion for the task of keyword veri¯cation. The reported experiments demonstrate that substantial improvements in veri¯cation performance can be obtained through the fusion of multiple keyword veri¯ers. The research focuses on combinations of speech background model based veri¯ers and cohort word veri¯ers. The ¯nal major contribution is a comprehensive study of the e®ects of limited training data for keyword spotting. This study is performed with consideration as to how these e®ects impact the immediate development and deployment of speech technologies for non-English languages.

Automatic spoken language identification utilizing acoustic and phonetic speech information

Automatic spoken Language Identi¯cation (LID) is the process of identifying the language spoken within an utterance. The challenge that this task presents is that no prior information is available indicating the content of the utterance or the identity of the speaker. The trend of globalization and the pervasive popularity of the Internet will amplify the need for the capabilities spoken language identi¯ca- tion systems provide. A prominent application arises in call centers dealing with speakers speaking di®erent languages. Another important application is to index or search huge speech data archives and corpora that contain multiple languages. The aim of this research is to develop techniques targeted at producing a fast and more accurate automatic spoken LID system compared to the previous National Institute of Standards and Technology (NIST) Language Recognition Evaluation. Acoustic and phonetic speech information are targeted as the most suitable fea- tures for representing the characteristics of a language. To model the acoustic speech features a Gaussian Mixture Model based approach is employed. Pho- netic speech information is extracted using existing speech recognition technol- ogy. Various techniques to improve LID accuracy are also studied. One approach examined is the employment of Vocal Tract Length Normalization to reduce the speech variation caused by di®erent speakers. A linear data fusion technique is adopted to combine the various aspects of information extracted from speech. As a result of this research, a LID system was implemented and presented for evaluation in the 2003 Language Recognition Evaluation conducted by the NIST.

Predictions of road traffic noise on residential balconies using a specular and diffusion model

A combined specular reflection and diffusion model using the radiosity technique was developed to calculate road traffic noise level on residential balconies. The model is capable of numerous geometrical configurations for a single balcony situated in the centre of a street canyon. The geometry of the balcony and the street can be altered with width,length and height. The model was used to calculate for three different geometrical and acoustic absorption characteristics for a balcony. The calculated results are presented in this paper.

Application of acoustic emission technology in monitoring structure integrity of bridges

Bridges are an important part of a nation’s infrastructure and reliable monitoring methods are necessary to ensure their safety and efficiency. Most bridges in use today were built decades ago and are now subjected to changes in load patterns that can cause localized distress, which can result in bridge failure if not corrected. Early detection of damage helps in prolonging lives of bridges and preventing catastrophic failures. This paper briefly reviews the various technologies currently used in health monitoring of bridge structures and in particular discusses the application and challenges of acoustic emission (AE) technology. Some of the results from laboratory experiments on a bridge model are also presented. The main objectives of these experiments are source localisation and assessment. The findings of the study can be expected to enhance the knowledge of acoustic emission process and thereby aid in the development of an effective bridge structure diagnostics system.

Analysis of acoustic emission data for accurate damage assessment for structural health monitoring applications

Structural health monitoring (SHM) refers to the procedure used to assess the condition of structures so that their performance can be monitored and any damage can be detected early. Early detection of damage and appropriate retrofitting will aid in preventing failure of the structure and save money spent on maintenance or replacement and ensure the structure operates safely and efficiently during its whole intended life. Though visual inspection and other techniques such as vibration based ones are available for SHM of structures such as bridges, the use of acoustic emission (AE) technique is an attractive option and is increasing in use. AE waves are high frequency stress waves generated by rapid release of energy from localised sources within a material, such as crack initiation and growth. AE technique involves recording these waves by means of sensors attached on the surface and then analysing the signals to extract information about the nature of the source. High sensitivity to crack growth, ability to locate source, passive nature (no need to supply energy from outside, but energy from damage source itself is utilised) and possibility to perform real time monitoring (detecting crack as it occurs or grows) are some of the attractive features of AE technique. In spite of these advantages, challenges still exist in using AE technique for monitoring applications, especially in the area of analysis of recorded AE data, as large volumes of data are usually generated during monitoring. The need for effective data analysis can be linked with three main aims of monitoring: (a) accurately locating the source of damage; (b) identifying and discriminating signals from different sources of acoustic emission and (c) quantifying the level of damage of AE source for severity assessment. In AE technique, the location of the emission source is usually calculated using the times of arrival and velocities of the AE signals recorded by a number of sensors. But complications arise as AE waves can travel in a structure in a number of different modes that have different velocities and frequencies. Hence, to accurately locate a source it is necessary to identify the modes recorded by the sensors. This study has proposed and tested the use of time-frequency analysis tools such as short time Fourier transform to identify the modes and the use of the velocities of these modes to achieve very accurate results. Further, this study has explored the possibility of reducing the number of sensors needed for data capture by using the velocities of modes captured by a single sensor for source localization. A major problem in practical use of AE technique is the presence of sources of AE other than crack related, such as rubbing and impacts between different components of a structure. These spurious AE signals often mask the signals from the crack activity; hence discrimination of signals to identify the sources is very important. This work developed a model that uses different signal processing tools such as cross-correlation, magnitude squared coherence and energy distribution in different frequency bands as well as modal analysis (comparing amplitudes of identified modes) for accurately differentiating signals from different simulated AE sources. Quantification tools to assess the severity of the damage sources are highly desirable in practical applications. Though different damage quantification methods have been proposed in AE technique, not all have achieved universal approval or have been approved as suitable for all situations. The b-value analysis, which involves the study of distribution of amplitudes of AE signals, and its modified form (known as improved b-value analysis), was investigated for suitability for damage quantification purposes in ductile materials such as steel. This was found to give encouraging results for analysis of data from laboratory, thereby extending the possibility of its use for real life structures. By addressing these primary issues, it is believed that this thesis has helped improve the effectiveness of AE technique for structural health monitoring of civil infrastructures such as bridges.

Development of a practical design guide for balcony acoustic treatments

Balcony acoustic treatments can be demonstrated to provide important benefits in reducing road traffic noise within the balcony space and consequently internally for any adjacent room. The actual effect on road traffic noise is derived from a multitude of variables that can be broadly categorized into (a) acoustical and (b) geometrical for two distinct propagation volumes being (i) the street space, and (ii) the balcony space. A series of recent research activities in this area has incorporated the use of a combined image and diffuse source model, which can be used to predict the effect of balconies on road traffic noise for large number of scenarios. This paper investigates and presents a method and capability to summarize predictive data into user friendly guidelines aimed for use by acoustical professionals and architects and possible implementation in building design policies for environmental noise. The paper concludes with a presentation of the likely format of a potential design guide.

Non-invasive clinical measurement of the viscoelastic properties of tendon using acoustic wave transmission

Background. In isotropic materials, the speed of acoustic wave propagation is governed by the bulk modulus and density. For tendon, which is a structural composite of fluid and collagen, however, there is some anisotropy requiring an adjustment for Poisson's ratio. This paper explores these relationships using data collected, in vivo, on human Achilles tendon and then compares estimates of elastic modulus and hysteresis against published values from in vitro mechanical tests. Methods. Measurements using conventional B-model ultrasound imaging, inverse dynamics and acoustic transmission techniques were used to determine dimensions, loading conditions and longitudinal speed of sound in the Achilles tendon during a series of isometric plantar flexion exercises against body weight. Upper and lower bounds for speed of sound versus tensile stress in the tendon were then modelled and estimates of the elastic modulus and hysteresis of the Achilles tendon derived. Results. Axial speed of sound varied between 1850 and 2090 ms-1 with a non-linear, asymptotic dependency on the level of tensile stress (5-35 MPa) in the tendon. Estimates derived for the elastic modulus of the Achilles tendon ranged between 1-2 GPa. Hysteresis derived from models of the stress-strain relationship, ranged from 3-11%. Discussion. Estimates of elastic modulus agree closely with those previously reported from direct measurements obtained via mechanical tensile tests on major weight bearing tendons in vitro [1,2]. Hysteresis derived from models of the stress-strain relationship is consistent with direct measures from various mamalian tendon (7-10%) but is lower than previous estimates in human tendon (17-26%) [3]. This non-invasive method would appear suitable for monitoring changes in tendon properties during dynamic sporting activities.

Simulating the effect of acoustic treatment types for residential balconies with road traffic noise

Development of design guides to estimate the difference in speech interference level due to road traffic noise between a reference position and balcony position or façade position is explored. A previously established and validated theoretical model incorporating direct, specular and diffuse reflection paths is used to create a database of results across a large number of scenarios. Nine balcony types with variable acoustic treatments are assessed to provide acoustic design guidance on optimised selection of balcony acoustic treatments based on location and street type. In total, the results database contains 9720 scenarios on which multivariate linear regression is conducted in order to derive an appropriate design guide equation. The best fit regression derived is a multivariable linear equation including modified exponential equations on each of nine deciding variables, (1) diffraction path difference, (2) ratio of total specular energy to direct energy, (3) distance loss between reference position and receiver position, (4) distance from source to balcony façade, (5) height of balcony floor above street, (6) balcony depth, (7) height of opposite buildings, (8) diffusion coefficient of buildings, and; (9) balcony average absorption. Overall, the regression correlation coefficient, R2, is 0.89 with 95% confidence standard error of ±3.4 dB.

Ion-acoustic waves in a complex plasma with negative ions

Negative ions and negatively charged micro- to nano-meter sized dust grains are ubiquitous in astrophysical as well as industrial processing plasmas. The negative ions can appear in electro-negative plasmas as a result of elementary processes such as dissociative or non-dissociative electron attachment to neutrals. They are usually rather small in number, and in general do not affect the overall plasma behavior. On the other hand, since the dust grains are almost always highly negative, even in small numbers they can take up a considerable proportion of the total negative charge in the system. The presence of dusts can affect the characteristics of most collective processes of the plasma since the charge balance in both the steady and dynamic states can be significantly altered. Another situation that often occurs is that the electron number density becomes small because of their absorption by the dust grains or the discharge walls. In this case the negative ions in the plasma can play a very important role. Here, a self-consistent theory of linear waves in complex laboratory plasmas containing dust grains and negative ions is presented. A comprehensive model for such plasmas including source and sink effects associated with the presence of dust grains and negative ions is introduced. The stationary state of the plasma as well as the dispersion and damping characteristics of the waves are investigated. All relevant processes, such as ionization, diffusion, electron attachment, negative-positive ion recombination, dust charge relaxation, and dissipation due to electron and ion elastic collisions with neutrals and dust particles, as well as charging collisions with the dusts, are taken into consideration.

Investigations on road noise level spatial variability within a specially designed acoustic balcony

An investigation into the spatial distribution of road traffic noise levels on a balcony is conducted. A balcony constructed to a special acoustic design due to its elevation above an 8 lane motorway is selected for detailed measurements. The as-constructed balcony design includes solid parapets, side walls, ceiling shields and highly absorptive material placed on the ceiling. Road traffic noise measurements are conducted spatially using a five channel acoustic analyzer, where four microphones are located at various positions within the balcony space and one microphone placed outside the parapet at a reference position. Spatial distributions in both vertical and horizontal planes are measured. A theoretical model and prediction configuration is presented that assesses the acoustic performance of the balcony under existing traffic flow conditions. The prediction model implements a combined direct path, specular reflection path and diffuse reflection path utilizing image source and radiosity techniques. Results obtained from the prediction model are presented and compared to the measurement results. The predictions are found to correlate well with measurements with some minor differences that are explained. It is determined that the prediction methodology is acceptable to assess a wider range of street and balcony configuration scenarios.

Primary and secondary neural networks of auditory prepulse inhibition: a functional magnetic resonance imaging study of sensorimotor gating of the human acoustic startle response

Feedforward inhibition deficits have been consistently demonstrated in a range of neuropsychiatric conditions using prepulse inhibition (PPI) of the acoustic startle eye-blink reflex when assessing sensorimotor gating. While PPI can be recorded in acutely decerebrated rats, behavioural, pharmacological and psychophysiological studies suggest the involvement of a complex neural network extending from brainstem nuclei to higher order cortical areas. The current functional magnetic resonance imaging study investigated the neural network underlying PPI and its association with electromyographically (EMG) recorded PPI of the acoustic startle eye-blink reflex in 16 healthy volunteers. A sparse imaging design was employed to model signal changes in blood oxygenation level-dependent (BOLD) responses to acoustic startle probes that were preceded by a prepulse at 120 ms or 480 ms stimulus onset asynchrony or without prepulse. Sensorimotor gating was EMG confirmed for the 120-ms prepulse condition, while startle responses in the 480-ms prepulse condition did not differ from startle alone. Multiple regression analysis of BOLD contrasts identified activation in pons, thalamus, caudate nuclei, left angular gyrus and bilaterally in anterior cingulate, associated with EMGrecorded sensorimotor gating. Planned contrasts confirmed increased pons activation for startle alone vs 120-ms prepulse condition, while increased anterior superior frontal gyrus activation was confirmed for the reverse contrast. Our findings are consistent with a primary pontine circuitry of sensorimotor gating that interconnects with inferior parietal, superior temporal, frontal and prefrontal cortices via thalamus and striatum. PPI processes in the prefrontal, frontal and superior temporal cortex were functionally distinct from sensorimotor gating.

Computer-assisted sampling of acoustic data for more efficient determination of bird species richness

Bird species richness survey is one of the most intriguing ecological topics for evaluating environmental health. Here, bird species richness denotes the number of unique bird species in a particular area. Factors affecting the investigation of bird species richness include weather, observation bias, and most importantly, the prohibitive costs of conducting surveys at large spatiotemporal scales. Thanks to advances in recording techniques, these problems have been alleviated by deploying sensors for acoustic data collection. Although automated detection techniques have been introduced to identify various bird species, the innate complexity of bird vocalizations, the background noise present in the recording and the escalating volumes of acoustic data pose a challenging task on determination of bird species richness. In this paper we proposed a two-step computer-assisted sampling approach for determining bird species richness in one-day acoustic data. First, a classification model is built based on acoustic indices for filtering out minutes that contain few bird species. Then the classified bird minutes are ordered by an acoustic index and the redundant temporal minutes are removed from the ranked minute sequence. The experimental results show that our method is more efficient in directing experts for determination of bird species compared with the previous methods.