Archives of Acoustics

Sound and vibrations are often perceived via the auditory and tactile senses simultaneously, e.g., in a car or train. During a rock concert, the body vibrates with the rhythm of the music. Even in a concert hall or a church, sound can excite vibrations in the ground or seats. These vibrations might not be perceived separately because they integrate with the other sensory modalities into one multi-modal perception.

This paper discusses the relation between sound and vibration for frequencies up to 1 kHz in an opera house and a church. Therefore, the transfer function between sound pressure and acceleration was measured at different exemplary listening positions. A dodecahedron loudspeaker on the stage was used as a sound source. Accelerometers on the ground, seat and arm rest measured the resulting vibrations. It was found that vibrations were excited over a broad frequency range via airborne sound. The transfer function was measured using various sound pressure levels. Thereby, no dependence on level was found. The acceleration level at the seat corresponds approximately to the sound pressure level and is independent of the receiver position. Stronger differences were measured for vibrations on the ground.

As a consequence of recent implementations of EU Directives related to noise protection more and more students of various AGH-UST programs are introduced to the basics of acoustic measurements. Students at various levels of theoretical background in the field of acoustic measurements are offered practical training in measurements using digital sound analyzers. The situation would be optimal if each student could have a device at his/her own disposal. Unfortunately, such a situation is not possible at the moment because of various reasons.

With the above problem in mind, a dedicated software package has been developed, implemented in the LabVIEW environment, which allows detailed studies of problems related to the acoustic signal measurement using sound level meters, as well as tasks in spectral analysis (1/1 and 1/3 band filters) and narrow-band (FFT) analysis. With such organization during the introductory laboratory classes each student is offered a direct individual contact with a virtual device that is properly pre-programmed for realization of a well-constructed learning process. It definitely facilitates understanding of the essence of acoustic signal measurements and provides a good basis for further laboratory work carried out as a team-activity.

We present a highly efficient filter structure to create power-complementary filter pairs for phantom source widening. It either introduces frequency-dependent phase or amplitude differences in a pair of loudspeaker signals. We evaluate how the perceptual effect is influenced by off-center listening positions in a standard ±30° loudspeaker setup. The evaluation of the phantom source widening effect is based on measurements of the inter-aural cross-correlation coefficient (IACC), which is justified by its pronounced correlation to the perceived phantom source width in prior listening test results.

The Gaussian mixture model (GMM) method is popular and efficient for voice conversion (VC), but it is often subject to overfitting. In this paper, the principal component regression (PCR) method is adopted for the spectral mapping between source speech and target speech, and the numbers of principal components are adjusted properly to prevent the overfitting. Then, in order to better model the nonlinear relationships between the source speech and target speech, the kernel principal component regression (KPCR) method is also proposed. Moreover, a KPCR combined with GMM method is further proposed to improve the accuracy of conversion. In addition, the discontinuity and oversmoothing problems of the traditional GMM method are also addressed. On the one hand, in order to solve the discontinuity problem, the adaptive median filter is adopted to smooth the posterior probabilities. On the other hand, the two mixture components with higher posterior probabilities for each frame are chosen for VC to reduce the oversmoothing problem. Finally, the objective and subjective experiments are carried out, and the results demonstrate that the proposed approach shows greatly better performance than the GMM method. In the objective tests, the proposed method shows lower cepstral distances and higher identification rates than the GMM method. While in the subjective tests, the proposed method obtains higher scores of preference and perceptual quality.

Reverberant responses are widely used to characterize acoustic properties of rooms, such as the early decay time (EDT) and the reverberation times T20 and T30. However, in real conditions a sound decay is often deformed by background noise, thus a precise evaluation of decay times from noisy room responses is the main problem. In this paper this issue is examined by means of numerical method where the decay times are estimated from the decay function that has been determined by nonlinear polynomial regression from a pressure envelope obtained via the discrete Hilbert transform. In numerical experiment the room responses were obtained from simulations of a sound decay for two-room coupled system. Calculation results have shown that background noise slightly affects the evaluation of reverberation times T20 and T30 as long as the signal-to-noise ratio (SNR) is not smaller than about 25 and 35 dB, respectively. However, when the SNR is close to about 20 and 30 dB, high overestimation of these times may occur as a result of bending up of the decay curve during the late decay.

There are an increasing number of binaural systems embedded with head-related transfer functions (HRTFs), so listeners can experience virtual environments via conventional stereo loudspeakers or head- phones. As HRTFs vary from person to person, it is difficult to select appropriated HRTFs from already existing databases for users. Once the HRTFs in a binaural audio device hardly match the real ones of the users, poor localization happens especially on the cone of confusion. The most accurate way to obtain personalized HRTFs might be doing practical measurements. It is, however, expensive and time consuming. Modifying non-individualized HRTFs may be an effort-saving way, though the modifications are always accompanied by undesired audio distortion. This paper proposes a flexible HRTF adjustment system for users to define their own HRTFs. Also, the system can keep sounds from suffering intolerable distortion based on an objective measurement tool for evaluating the quality of processed audio.

Filled pauses (FPs) have proved to be more than valuable cues to speech production processes and important units in discourse analysis. Some aspects of their form and occurrence patterns have been shown to be speaker- and language-specific. In the present study, basic acoustic properties of FPs in Polish task-oriented dialogues are explored. A set of FPs was extracted from a corpus of twenty task- oriented dialogues on the basis of available annotations. After initial scrutiny and selection, a subset of the signals underwent a series of pitch, formant frequency and voice quality analyses. A significant amount of variation found in the realisations of FPs justifies their potential application in speaker recognition systems. Regular monosegmental FPs were confirmed to show relatively stable basic acoustic parameters, which allows for their easy identification and measurements but it may result in less significant differences among the speakers.

Two violins were investigated. The only intentionally introduced difference between them was the type of varnish. One of the instruments was covered with a spirit varnish, the other was oil varnished. Experimental modal analysis was done for unvarnished/varnished violins and a questionnaire inquiry on the instrument’s sound quality was performed. The aim of both examinations was to find differences and similarities between the two instruments in the objective (modal parameters) and subjective domain (subjective evaluation of sound quality). In the modal analysis, three strongly radiating signature modes were taken into account. Varnishing did not change the sequence of mode shapes. Modal frequencies A0 and B(1+) were not changed by oil varnishing compared to the unvarnished condition. For the oil varnished instrument, the frequency of mode B(1+) was lower than that of the same mode of the spirit varnished instrument. Our two violins were not excellent instruments, but before varnishing they were practically identical. However, after varnishing it appeared that the oil-varnished violin was better than the spirit-varnished instrument. Therefore, it can be assumed with a fairly high probability that also in general, the oil-varnished violins sound somewhat better than initially identical spirit-varnished ones.