Yoshio Yamasaki

Sound field measurements based on reconstruction from laser projections

In this paper, we describe some new sound field measurement methods by using a laser Doppler vibrometer (LDV). By irradiating the reflection wall with a laser, we can observe the light velocity change that is caused by the refractive index change from the change in air density. It means that it is possible to observe the change of the sound pressure. We measured a sound field projection on a 2D plane using a scanning laser Doppler vibrometer (SVM) which can visualize a sound field. And we made a 3D sound field reconstruction from some 2D laser projections based on computed tomography (CT) techniques. We made the reconstructed image for the sound field near the loudspeaker or in the room.

Measurement of spatial information in sound fields by a closely located four‐point microphone method

When an estimation is made of the sound field in a room, it is important to grasp spatial information, especially for the early reverberation periods. Here, a way to grasp spatial information of sound fields from impulse responses measured at four closely located points, the origin, and three points the same distance (3∼5 cm) from the origin on the rectangular coordinate axes will be discussed. From these four impulse responses the position of virtual image sources are calculated by the correlation and/or intensity technique. Concert halls in Europe, the United States, and Japan and some scale models are measured by this method. The three‐dimensional distribution of virtual image sources, directivity patterns, and impulse responses from several directions are shown. Calculation of signals from special directions by convolving these impulse responses with dry music to estimate the sound fields with ears or to help the acoustic design, and the introduction of the Wignet distribution are also discussed.

Effect of transient signal length on cross‐correlation functions in a room

An investigation was made of cross‐correlation functions with transient signals between two points, which correspond to both ears of a listener in a room. The interaural cross correlation is closely related to the subjective impressions of sound fields. The study attempted to account in a comprehensive way for the combined effects that initial reflected and reverberation sounds from music or other transient signals have on such impressions. To this end, cross‐correlation functions of the rise and fall of the sound field from transient signals were derived from the impulse responses at two points in the hall. These results were combined with image‐sources distribution patterns derived by the closely located four‐point‐microphone method; then, a comparative explanation was made of the changes with transient time duration of the cross‐correlation functions. Good agreement was found between changes with time in experimentally derived maximal cross‐correlation function values and the changes with time in image...

An application of digital techniques to obtain the spatial information of a room

The impulse response represents the transmission characteristics from the certain source point to the receiving point of the room, if we consider it to be a linear system. According to the position of the source and receiving points some differences are caused in the impulse responses, but they imply the spatial characteristics of the room. The mirror image method is often used to obtain the spatial characteristics of a room, but it is difficult to apply this method to an actual room because of the complexity of the room shape. At four slightly variant receiving points on x, y, and z axis, we measured the impulse responses of the same source point. Then we computed the image sources to the three dimensional field. From these virtual image sources, we obtained SPL distributions, conditions of standing waves, effects of absorbing materials, mean free path, etc., in a room.

Speaker Verification Using Narrow-band Envelope Correlation Matrices

We confirmed that a speakers vocal individuality is contained in the inter-band correlations of narrow-band (1/4 or 1/8 octave bands) temporal envelopes. Two types of envelope correlation matrices (ECMs) were made for 53 speakers, using three utterances of an identical sentence (assuming a situation where a password for verification was stolen) so that any differences in the spoken contents might not greatly influence their individuality. Type-A (reference) ECMs of two of the utterances were constructed to make a speakers individual template, and a type-B ECM was constructed using the other utterance. Speaker matching tests between the two types of ECMs, based on Gaussian mixture model (GMM) matching scores, verified the validity of the individual speakers. In particular, a speakers voice could be verified using spoken materials through the telephone band (250 Hz 3 kHz), a high frequency range (2- 11.3 kHz), or a wide frequency range (250 Hz - 11.3 kHz)

Preference test of seven European concert halls

Preference tests of seven European concert halls using musical motifs recorded through a dummy head were performed. Test signals were reproduced binaurally. Eighty‐eight students served as subjects. The results show that (1) there was no significant difference between preference scores for these halls when averaged over all subjects, and (2) subjects could be divided into several groups with respect to preference. In conclusion, even though each subject was free to prefer hall A to hall B, the acoustical quality of a certain hall could not be solely evaluated using the preference score. [Work supported, in part, by The Kajima Foundation.]

Communication aid utilizing bone-conducted sound via teeth by means of mouthpiece form actuator

Since bone-conducted sound is conveyed to cochlea directly, without passing through eardrum, it is audible even for hard-of-hearing people whose inner ears are still normal. In this study, we utilize bone-conducted sound via teeth so as to support sound communication. We implement a bone-conducted actuator on a tooth, while actuators of prevalent hearing aids are attached to mastoid, forehead, or jaw in general. Teeth convey sound excitation more easily, because they are bare bones, not covered with skin. Our hearing aid is made in the form of mouthpiece, and it can be readily put on and taken off from a tooth. Plus, we carry out experiments regarding sound localization and thresholds of bone-conducted sound via teeth, using this actuator. The results offer hearing characteristic of bone-conducted sound via teeth and show that examinees can perceive right and left using bone-conducted sound via teeth. In addition, we also attempt to record vibrations of teeth through a microphone, which is embedded on the mouthpiece form actuator. The aim of this study is to realize a hybrid actuator that enables both hearing and recording simultaneously and to suggest a new communication aid system not only for hard-of-hearing people but also for the robust.

Extract voice information using high-speed camera

Conversation is one of the most important channels for human beings. To help communications, speech recognition technologies have been developed. Above all, in a conversation, not only contents of utterances but also intonations and tones include important information regarding a speakers intention. To study the sphere of human speech, microphones are typically used to record voices. However, since microphones have to be set around a space, their existences affect a physical behavior of the sound field. To challenge this problem, we have suggested a recording method using a high-speed camera. By using a high-speed camera for recording sound vibrations, it can record two or more points within the range of the camera at the same time and can record from a distance, without interfering with the sound fields. In this study, we extract voice information using high-speed videos which capture both a face and a cervical part of the subject. This method allows recording skin vibrations which contain voices with i...

Design of resonant frequencies for piezoelectric actuator with integrated components

Piezoelectric actuators are used in a wide range of electrical devices, including piezoelectric speakers, buzzers, haptics and ultrasonic transducers. For piezoelectric actuator systems used in mobile devices, the most important issue is improving the electromechanical conversion efficiency. The power consumed by the actuators must be minimized due to the small size of the batteries used. The frequency response around the mechanical resonance must be carefully designed to enable low power driving. The resonant frequencies of piezoelectric actuators that consist of integrated components, such as the metal cones in ultrasonic speakers, are determined by the energy dispersion of the total system. Therefore, factors such as the size and physical properties of each component must be designed to optimize the resonant frequencies for practical applications. The total energy of the piezoelectric system is described by Lagrange-Maxwell equations. Even though it is not easy to solve the differential equations writt...

High‐efficiency, high‐speed, 1‐bit coding using generalized harmonic analysis

Usually in high‐efficiency coding such as MPEG a regular digital signal whose sampling rate is from 32 to 48 kHz, and number of quantification bits from 12 to 16 bits are used. High‐speed, 1‐bit coding is proposed. Generalized harmonic analysis is simple. A sinusoid that minimizes the power of the residual waveform is found and extracted from the original waveform. The same procedure is carried out by substituting the weakest residual waveform for the original waveform. The frequency component can be extracted correctly. High‐efficiency, high‐speed, 1‐bit coding using generalized harmonic analysis is reported here.