Recent advances in ubiquitous technology have greatly changed the way people stay connected. We conducted an in-depth video shadowing study to observe how close-knit groups use all the technology at their disposal to stay in touch and share their lives. We observed a pattern of related behaviors that we call channel blending, the integration of interactions and content over multiple channels into one coherent conversation, often including both local and remote participants. Channel blending is the opposite of multitasking in that it involves merging many lines of focus into one, rather than switching attention between them. We discuss ways technology could better support this emerging style of multichannel content-sharing and communication.

In recent years Computer-Assisted Language Learning (CALL) systems have been widely used in foreign language education. Some systems use automatic speech recognition (ASR) technologies to detect pronunciation errors and estimate the proficiency level of individual students. When speech recording is done in a CALL classroom, however, utterances of a student are always recorded with those of the others in the same class. The latter utterances are just background noise, and the performance of automatic pronunciation assessment is degraded especially when a student is surrounded with very active students. To solve this problem, we apply a noise reduction technique, Stereo-based Piecewise Linear Compensation for Environments (SPLICE), and the compensated feature sequences are input to a Goodness Of Pronunciation (GOP) assessment system. Results show that SPLICE-based noise reduction works very well as a means to improve the assessment performance in a noisy classroom.

In this paper, we propose a novel speaker adaptation technique, regularized-MLLR, for Computer Assisted Language Learn-ing (CALL) systems. This method uses a linear combination of a group of teachers ’ transformation matrices to represent each target learner’s transformation matrix, thus avoids the over-adaptation problem that erroneous pronunciations come to be judged as good pronunciations after conventional MLLR speaker adaptation, which uses learners ’ “imperfect ” speech as target utterances of adaptation. Experiments of automatic scor-ing and error detection on public databases show that the pro-posed method outperforms conventional MLLR adaption in pronunciation evaluation and can avoid the problem of over adaptation. Index Terms: Computer Assisted Language Learning (CALL), speaker adaption, pronunciation evaluation, goodness of pro-nunciation (GOP), maximum likelihood linear regression (MLLR) 1.

In this paper, we investigate the effects and problems of MLLR speaker adaptation when applied to pronunciation evaluation. Automatic scoring and error detection experiments are conducted on two publicly available databases of Japanese learners’ English pronunciation. As we expected, overadaptation causes misjudge of pronunciation accuracy. Following these experiments, two novel methods, Forced-aligned GOP scoring and Regularized-MLLR adaptation, are proposed to solve the adverse effects of MLLR adaption. Experimental results show that the proposed methods can better utilize MLLR adaptation and avoid over-adaptation. Index Terms: Computer Assisted Language Learning (CALL), speaker adaption, pronunciation evaluation, goodness of pronunciation (GOP), maximum likelihood linear regression (MLLR)

In foreign language learning, shadowing has been used as a method for improving speaking and listening ability. In this method, learners are required to repeat a presented native utterance as closely and quickly as possible. Since learners have to follow the speaking rate of the presented utterance, their pronunciation often becomes very inarticulate and unintelligible. These features of shadowing make it very difficult to build a reliable scoring system for shadowing productions. In this paper, two techniques are proposed and investigated for automatic scoring of shadowing productions. Experiments show that good correlations are found between automatic scores and TOEIC overall proficiency scores. Index Terms: shadowing, automatic scoring, articulatory effort, goodness of pronunciation, bottom-up clustering

In our ethnographic investigation of software integration projects a recurrent pattern emerges. The detailed understanding leaders have of the design and development decreases over time as they become busier and busier attending meetings, creating documents, and resolving issues and thus cannot spend much time on design or development work. As a result, their leadership becomes increasingly decoupled from the work of the project. We discuss various dimensions of this problem.

A program for self assessment of Japanese pronunciation by Englishspeaking learners was developed using a language model built with input from a language teacher in collaboration with speech engineers. This collaboration enhanced the program's capacity for accurate assessment and provides practical support to users by linking evaluation with feedback, and an editorial function of error patterns. The program drew positive responses from participants in a trial run. This paper discusses the development of our language model, the function and evaluation of this self assessment program.

Service technicians in the field often come across difficult service problems that are new to them. They have a large number of resources that they can draw on to deal with such problems, including both people and documents. We have undertaken a detailed study of technicians' everyday work, and have discovered two distinct types of information use, reflecting two different problem-solving practices. The less frequently used problem-solving practice is instruction following, where technicians follow company-documented Repair Analysis Procedures (RAPs). The second, more common practice is gleaning, where the information is gathered from many sources -- including other technicians and informal tips, which are documents written by technicians describing their invented solutions to hard service problems. Our observations show how the informational and interface affordances of the system for accessing the tips support their easy incorporation into the gleaning approach for problem solving in difficult cases. We also recommend ways that RAPs can be augmented to provide affordances for gleaning, and more effective instruction following.

Open-source software, usually created by volunteer programmers dispersed worldwide, now competes with that developed by software firms. This achievement is particularly impressive as open-source programmers rarely meet. They rely heavily on electronic media, which preclude the benefits of face-to-face contact that programmers enjoy within firms. In this paper, we describe findings that address this paradox based on observation, interviews and quantitative analyses of two open-source projects. The findings suggest that spontaneous work coordinated afterward is effective, rational organizational culture helps achieve agreement among members and communications media moderately support spontaneous work. These findings can imply a new model of dispersed collaboration.

A new approach for developing an intelligent massage robot is presented. Massage is popular as a form of body conditioning and was been mechanized in our country. However, conventional machines have some drawbacks to be improved on providing a certain comfort close to human therapies. We focus on ways to realize kneading massage actions which are considered to be the most difficult for machines. The action is first shown in a physical manner and the control is discussed based on a position/force hybrid controller. Then, some related problems are made apparent which have led to a new concept using a learning or adaptive mechanism. Finally, experiments using a test bed called the MTU (mechanotherapy unit) have proven that there is a good possibility of developing an innovative massage robot.