Toshiaki Sugimura

Electronic device, vibration generator, vibration-type reporting method, and report control method

An electronic device drives an oscillatory actuator to cause generation of vibration when it is detected that an operation input to a touch panel or operation key has been received. The electronic device causes, by this vibration, the touch panel and operation key to vibrate in a direction perpendicular to their respective front surfaces. Alternatively, the housing of the electronic device is made to vibrate. Further, the oscillatory actuator has a weight, a support member for supporting the weight to allow it to reciprocate, and connected to the touch panel or housing or other vibratory member of the electronic device, or vibratory member of the base member of the oscillatory actuator in contact with the vibratory member, and a mechanism for imparting a magnetic force or electrostatic force to cause reciprocal movement of the weight.

A tree-structured clustering method integrating noise and SNR for piecewise linear-transformation-based noise adaptation

This paper proposes the application of a tree-structured clustering method that integrates the effects of noise as well as SNR variation in the framework of piecewise-linear transformation (PLT)-based noise adaptation for robust speech recognition. According to the clustering results, a noisy speech HMM is made for each node of the tree structure. An HMM that best matches the input speech is selected based on the likelihood maximization criterion by tracing the tree downward from the top (root), and the selected HMM is further adapted by linear transformation. The proposed method is evaluated by applying it to a Japanese dialogue recognition system. Experimental results confirm that the proposed method is effective in recognizing numerically noise-added speech and actual noisy speech uttered by a wide range of speakers under various noise conditions.

Seamless Service Handoff for Ubiquitous Mobile Multimedia

To realize a ubiquitous multimedia environments utilizing mobile devices with multimedia appliances over heterogeneous access networks, we design and implement a service handoff (HO) system which supports two kinds of mobility; vertical HO between 3G Cellular and wireless LAN, and media redirection from mobile devices to multimedia applicances. To overcome inherently high setup latency of Cellular and multimedia appliances, which causes severe media disruption and throughput degradation, we propose a proactive service HO method with a soft HO mechanism based on a geographical resource map. Location-aided HO using the map allows the system to initiate HO in advance for concealment of setup latency. We evaluate HO performance on a testbed.

Tree-Structured Clustering Methods for Piecewise Linear-Transformation-Based Noise Adaptation

This paper proposes the application of tree-structured clustering to the processing of noisy speech collected under various SNR conditions in the framework of piecewise-linear transformation (PLT)-based HMM adaptation for noisy speech. Three kinds of clustering methods are described: a one-step clustering method that integrates noise and SNR conditions and two two-step clustering methods that construct trees for each SNR condition. According to the clustering results, a noisy speech HMM is made for each node of the tree structure. Based on the likelihood maximization criterion, the HMM that best matches the input speech is selected by tracing the tree from top to bottom, and the selected HMM is further adapted by linear transformation. The proposed methods are evaluated by applying them to a Japanese dialogue recognition system. The results confirm that the proposed methods are effective in recognizing digitally noise-added speech and actual noisy speech issued by a wide range of speakers under various noise conditions. The results also indicate that the one-step clustering method gives better performance than the two-step clustering methods.

A ring-shaped EMG measurement system for applying to user interface

This paper proposes a ring-shaped EMG measurement system to be worn on one or more fingers for a new user interface that relies on EMG measurements. A preamplifier is mounted on each electrode to suppress the effect of noise. The system is truly wearable since the main amplifier is designed to be worn on the wrist. The proposed EMG measurement system is implemented and evaluated. The experiments confirm that the proposed electrode is superior to conventional electrodes in terms of stability and usability while offering excellent long term performance. The effectiveness of the proposed electrode system is confirmed by the test in which the interface realizes highly accurate speech recognition via EMG.

Robust telop character recognition in video for content-based retrieval

The recognition of telop characters in video has two problems: edge degradation and background noise. To overcome these problems, the paper proposes: (1) a feature that describes the shape of background region in addition to character region; (2) a classifier that dynamically decreases the influence of noise using the difference in the number of pixels within a local area. Experiments show that the two proposals achieve a recognition rate of 76% and a cumulative classification rate within the best ten of 92%. Both values are significantly better than those of conventional methods.

Life size humanoid robot that reproduces gestures as a communication terminal: appearance considerations

We propose an application that uses a humanoid robot as a communications terminal. We describe a life-size robot that suits inter-personal communication. For simplicity, the current version is restricted to a motorized chair. We show that the robot realizes three service modes: chat mode, e-mail mode and autonomous mode. In chat mode, the hand gestures of the humanoid robot reproduce those made by the operator as captured by surface EMG signal recognition. the humanoid robot is seen as a feasible instrument of telecommunication.