Audio visual automatic speech recognition in vehicles

Acoustically, car cabins are extremely noisy and as a consequence, existing audio-only speech recognition systems, for voice-based control of vehicle functions such as the GPS based navigator, perform poorly. Audio-only speech recognition systems fail to make use of the visual modality of speech (eg: lip movements). As the visual modality is immune to acoustic noise, utilising this visual information in conjunction with an audio only speech recognition system has the potential to improve the accuracy of the system. The field of recognising speech using both auditory and visual inputs is known as Audio Visual Speech Recognition (AVSR). Continuous research in AVASR field has been ongoing for the past twenty-five years with notable progress being made. However, the practical deployment of AVASR systems for use in a variety of real-world applications has not yet emerged. The main reason is due to most research to date neglecting to address variabilities in the visual domain such as illumination and viewpoint in the design of the visual front-end of the AVSR system. In this paper we present an AVASR system in a real-world car environment using the AVICAR database [1], which is publicly available in-car database and we show that the use of visual speech conjunction with the audio modality is a better approach to improve the robustness and effectiveness of voice-only recognition systems in car cabin environments.

Coding speech signals at very low rates (below 1 kb/s) with high intelligibility

This thesis presents an original approach to parametric speech coding at rates below 1 kbitsjsec, primarily for speech storage applications. Essential processes considered in this research encompass efficient characterization of evolutionary configuration of vocal tract to follow phonemic features with high fidelity, representation of speech excitation using minimal parameters with minor degradation in naturalness of synthesized speech, and finally, quantization of resulting parameters at the nominated rates. For encoding speech spectral features, a new method relying on Temporal Decomposition (TD) is developed which efficiently compresses spectral information through interpolation between most steady points over time trajectories of spectral parameters using a new basis function. The compression ratio provided by the method is independent of the updating rate of the feature vectors, hence allows high resolution in tracking significant temporal variations of speech formants with no effect on the spectral data rate. Accordingly, regardless of the quantization technique employed, the method yields a high compression ratio without sacrificing speech intelligibility. Several new techniques for improving performance of the interpolation of spectral parameters through phonetically-based analysis are proposed and implemented in this research, comprising event approximated TD, near-optimal shaping event approximating functions, efficient speech parametrization for TD on the basis of an extensive investigation originally reported in this thesis, and a hierarchical error minimization algorithm for decomposition of feature parameters which significantly reduces the complexity of the interpolation process. Speech excitation in this work is characterized based on a novel Multi-Band Excitation paradigm which accurately determines the harmonic structure in the LPC (linear predictive coding) residual spectra, within individual bands, using the concept 11 of Instantaneous Frequency (IF) estimation in frequency domain. The model yields aneffective two-band approximation to excitation and computes pitch and voicing with high accuracy as well. New methods for interpolative coding of pitch and gain contours are also developed in this thesis. For pitch, relying on the correlation between phonetic evolution and pitch variations during voiced speech segments, TD is employed to interpolate the pitch contour between critical points introduced by event centroids. This compresses pitch contour in the ratio of about 1/10 with negligible error. To approximate gain contour, a set of uniformly-distributed Gaussian event-like functions is used which reduces the amount of gain information to about 1/6 with acceptable accuracy. The thesis also addresses a new quantization method applied to spectral features on the basis of statistical properties and spectral sensitivity of spectral parameters extracted from TD-based analysis. The experimental results show that good quality speech, comparable to that of conventional coders at rates over 2 kbits/sec, can be achieved at rates 650-990 bits/sec.

Dynamic visual features for audio-visual speaker verification

The cascading appearance-based (CAB) feature extraction technique has established itself as the state-of-the-art in extracting dynamic visual speech features for speech recognition. In this paper, we will focus on investigating the effectiveness of this technique for the related speaker verification application. By investigating the speaker verification ability of each stage of the cascade we will demonstrate that the same steps taken to reduce static speaker and environmental information for the visual speech recognition application also provide similar improvements for visual speaker recognition. A further study is conducted comparing synchronous HMM (SHMM) based fusion of CAB visual features and traditional perceptual linear predictive (PLP) acoustic features to show that higher complexity inherit in the SHMM approach does not appear to provide any improvement in the final audio-visual speaker verification system over simpler utterance level score fusion.

aa-cell in practice : an approach to musical live coding

Live coding performances provide a context with particular demands and limitations for music making. In this paper we discuss how as the live coding duo aa-cell we have responded to these challenges, and what this experience has revealed about the computational representation of music and approaches to interactive computer music performance. In particular we have identified several effective and efficient processes that underpin our practice including probability, linearity, periodicity, set theory, and recursion and describe how these are applied and combined to build sophisticated musical structures. In addition, we outline aspects of our performance practice that respond to the improvisational, collaborative and communicative requirements of musical live coding.

Recognising audio-visual speech in vehicles using the AVICAR database

Interacting with technology within a vehicle environment using a voice interface can greatly reduce the effects of driver distraction. Most current approaches to this problem only utilise the audio signal, making them susceptible to acoustic noise. An obvious approach to circumvent this is to use the visual modality in addition. However, capturing, storing and distributing audio-visual data in a vehicle environment is very costly and difficult. One current dataset available for such research is the AVICAR [1] database. Unfortunately this database is largely unusable due to timing mismatch between the two streams and in addition, no protocol is available. We have overcome this problem by re-synchronising the streams on the phone-number portion of the dataset and established a protocol for further research. This paper presents the first audio-visual results on this dataset for speaker-independent speech recognition. We hope this will serve as a catalyst for future research in this area.

Improving visual noise insensitivity in small vocabulary audio visual speech recognition applications

Visual noise insensitivity is important to audio visual speech recognition (AVSR). Visual noise can take on a number of forms such as varying frame rate, occlusion, lighting or speaker variabilities. The use of a high dimensional secondary classifier on the word likelihood scores from both the audio and video modalities is investigated for the purposes of adaptive fusion. Preliminary results are presented demonstrating performance above the catastrophic fusion boundary for our confidence measure irrespective of the type of visual noise presented to it. Our experiments were restricted to small vocabulary applications.

Can audio-visual speech recognition outperform acoustically enhanced speech recognition in automotive environment?

The use of visual features in the form of lip movements to improve the performance of acoustic speech recognition has been shown to work well, particularly in noisy acoustic conditions. However, whether this technique can outperform speech recognition incorporating well-known acoustic enhancement techniques, such as spectral subtraction, or multi-channel beamforming is not known. This is an important question to be answered especially in an automotive environment, for the design of an efficient human-vehicle computer interface. We perform a variety of speech recognition experiments on a challenging automotive speech dataset and results show that synchronous HMM-based audio-visual fusion can outperform traditional single as well as multi-channel acoustic speech enhancement techniques. We also show that further improvement in recognition performance can be obtained by fusing speech-enhanced audio with the visual modality, demonstrating the complementary nature of the two robust speech recognition approaches.

Engaging middle school students in context-based science : One teacher's approach

In Australia, there is a crisis in science education with students becoming disengaged with canonical science in the middle years of schooling. One recent initiative that aims to improve student interest and motivation without diminishing conceptual understanding is the context-based approach. Contextual units that connect the canonical science with the students’ real world of their local community have been used in the senior years but are new in the middle years. This ethnographic study explored the learning transactions that occurred in one 9th grade science class studying an Environmental Science unit for 11 weeks. Data were derived from field notes, audio and video recorded conversations, interviews, student journals and classroom documents with a particular focus on two selected groups of students. Data were analysed qualitatively through coding for emergent themes. This paper presents an outline of the program and discussion of three assertions derived from the preliminary analysis of the data. Firstly, an integrated, coherent sequence of learning experiences that included weekly visits to a creek adjacent to the school enabled the teacher to contextualise the science in the students’ local community. Secondly, content was predominantly taught on a need-to-know basis and thirdly, the lesson sequence aligned with a model for context-based teaching. Research, teaching and policy implications of these results for promoting the context-based teaching of science in the middle years are discussed.

Young adults’ accounts of scientific knowledge when responding to a television news report of contested science

This paper reports one aspect of a study of 28 young adults (18–26 years) engaging with the uncertain (contested) science of a television news report about recent research into mobile phone health risks. The aim of the study was to examine these young people’s ‘accounts of scientific knowledge’ in this context. Seven groups of friends responded to the news report, initially in focus group discussions. Later in semi-structured interviews they elaborated their understanding of the nature of science through their explanations of the scientists’ disagreement and described their mobile phone safety risk assessments. This paper presents their accounts in terms of their views of the nature of science and their concept understanding. Discussions were audio-recorded then analysed by coding the talk in terms of issues raised, which were grouped into themes and interpreted in terms of a moderate social constructionist theoretical framing. In this context, most participants expressed a ‘common sense’ view of the nature of science, describing it as an atheoretical, technical procedure of scientists testing their personal opinions on the issue, subject to the influence of funding sponsors. The roles of theory and data interpretation were largely ignored. It is argued that the nature of science understanding is crucial to engagement with contemporary socioscientific issues, particularly the roles of argumentation, theory, data interpretation, and the distinction of science from common sense. Implications for school science relate primarily to nature of science teaching and the inclusion of socioscientific issues in school science curricula. Future research directions are considered.

Measuring bitrate and quality trade-off in a fast region-of-interest based video coding

Prevailing video adaptation solutions change the quality of the video uniformly throughout the whole frame in the bitrate adjustment process; while region-of-interest (ROI)-based solutions selectively retains the quality in the areas of the frame where the viewers are more likely to pay more attention to. ROI-based coding can improve perceptual quality and viewer satisfaction while trading off some bandwidth. However, there has been no comprehensive study to measure the bitrate vs. perceptual quality trade-off so far. The paper proposes an ROI detection scheme for videos, which is characterized with low computational complexity and robustness, and measures the bitrate vs. quality trade-off for ROI-based encoding using a state-of-the-art H.264/AVC encoder to justify the viability of this type of encoding method. The results from the subjective quality test reveal that ROI-based encoding achieves a significant perceptual quality improvement over the encoding with uniform quality at the cost of slightly more bits. Based on the bitrate measurements and subjective quality assessments, the bitrate and the perceptual quality estimation models for non-scalable ROI-based video coding (AVC) are developed, which are found to be similar to the models for scalable video coding (SVC).

An audio wiki supporting mobile collaboration

Wikis have proved to be very effective collaboration and knowledge management tools in large variety of fields thanks to their simplicity and flexible nature. Another important development for the internet is the emergence of powerful mobile devices supported by fast and reliable wireless networks. The combination of these developments begs the question of how to extend wikis on mobile devices and how to leverage mobile devices' rich modalities to supplement current wikis. Realizing that composing and consuming through auditory channel is the most natural and efficient way for mobile device user, this paper explores the use of audio as the medium of wiki. Our work, as the first step towards this direction, creates a framework called Mobile Audio Wiki which facilitates asynchronous audio-mediated collaboration on the move. In this paper, we present the design of Mobile Audio Wiki. As a part of such design, we propose an innovative approach for a light-weight audio content annotation system for enabling group editing, versioning and cross-linking among audio clips. To elucidate the novel collaboration model introduced by Mobile Audio Wiki, its four usage modes are identified and presented in storyboard format. Finally, we describe the initial design for presentation and navigation of Mobile Audio Wiki.