Hisashi Takeshima

Equal‐loudness relations at high frequencies: Implications for loudness growth

Equal‐loudness relations spanning a wide range of sensation levels (SL) were measured by the method of adjustment for 12 pairs of frequencies from 1 to 16 kHz. Sensation levels were based on individual thresholds measured prior to the equal‐loudness determinations by an adaptive 2IFC procedure. To test for transitivity, three reference frequencies at 1, 3.15, and 5 kHz were used. Below 8 kHz, the matching results were well described by a linear function with a slope of 1.0; at higher frequencies, the functions became increasingly curvilinear. Near threshold and at moderate SLs the loudness of a tone at 12.5 kHz and higher increased more rapidly than the loudness of the lower‐frequency reference tone; at high SLs it increased more slowly. The dependence of the slope of the matching function on level was most pronounced at 16 kHz. Moreover, nearly perfect transitivity was observed. The reduction in slope at high SLs is consistent with recent evidence that severely restricting the spread of excitation to high frequencies decreases the tone’s rate of loudness growth [R. P. Hellman, J. Acoust. Soc. Am. 96, 2655–2663 (1994)]. [Work supported by the Japanese Ministry of Education, Science, Sports, and Culture.]

Estimation of the new equal‐loudness level contours

The current international standard of the equal‐loudness level contours specified in ISO 226 is found to involve large errors, epecially for frequencies below 1 kHz. In the past 10 years, a series of experiments has been conducted for full revision of ISO 226 in ISO/TC 43. At the final stage of this project, the new equal‐loudness level contours should be drawn from available data points. To do this, the use of an appropriate model for loudness perception is actually useful. A loudness function is proposed by combining that proposed by Lochner and Burger to express the steepness near threshold with the two‐stage model proposed by Attneave to consider the loudness‐comparison process. Equal‐loudness levels are then estimated according to the following procedure. (1) Parameters of the loudness function are estimated from the experimental data by the nonlinear least‐squares method. (2) The estimated parameters are smoothed along the frequency axis with B‐spline functions. (3) The equal‐loudness level contours...

Evaluation of steady noise from a multi-dimensional point of view

Abstract This paper describes the results of a study aimed at establishing a method of multi-dimensional noise evaluation. A description of the total negative impression of noise is obtained through subjective experiments with steady and almost steady sounds. The authors tried to determine Japanese descriptive adjectives independent of “ookii” (loud) in order to assess the total negative impression of noise. To do this, the authors introduced “degree of absolute noisiness” for evaluation of noise and measured it directly. Major conclusions are as follows: (1) at least two dimensions are needed to describe the total negative impression of noise; (2) one of the two dimensions is highly correlated with the Japanese adjective which means loud, and the other axis is correlated with the Japanese descriptive adjectives meaning harsh and unpleasant; (3) the scale “loud” correlates well with Zwickers loudness and the scale “harsh” correlates highly with Bismarcks sharpness.

The new International Organization for Standardization (ISO) standard for the equal‐loudness contours: Comparison to earlier contours and procedures

Equal‐loudness contours exhibit the intensity‐frequency characteristic of the auditory system. Since 1956 such contours based on the free‐field measurements of Robinson and Dadson [Br. J. Appl. Phys. 7, 166–181 (1956)] have been widely accepted as an international standard in ISO 226. However, in 1985 some questions about the general applicability of these standard contours were raised to ISO TC/43 WG1 by Fastl and Zwicker. Consequently, a new international effort to measure the equal‐loudness contours was undertaken. In 2003, this effort produced a revision of ISO 226. The newer experimental data and their analyses were subsequently published by Suzuki and Takeshima, [J. Acoust. Soc. Am. 116, 918–933 (2004)]. The revised contours more closely agree with the classic contours measured by Fletcher and Munson [J. Acoust. Soc. Am. 5, 82–108 (1933)] than with those produced by Robinson and Dadson. This finding is especially true below 500 Hz at moderate loudness levels. Moreover, the revised contours are in go...

Least‐mean‐square estimation of new equal‐loudness level contours from recent data based on a loudness perception model

Since probable large errors were suggested in 1985 in the equal‐loudness level contours by Robinson and Dadson [Br. J. Appl. Phys. 7, 166–181 (1956)], which were standardized as ISO 226, a considerable amount of data on the equal‐loudness relation has been accumulated. Most of the data consistently show a large discrepancy up to more than 20 phons from the contours, especially below 1 kHz. To obtain reliable contours from these new data sporadically given for some specific frequencies and phons, a model function representing the equal‐loudness relation was derived from a loudness function modified by the two‐stage loudness perception model. Values of the parameters of the model function were obtained by fitting the function to the experimental data. Equal‐loudness level contours could be drawn by use of the model function with the parameter values interpolated along the frequency axis. The resultant equal‐loudness level contours showed clear differences from those by Robinson and Dadson for all loudness l...

Equal‐loudness contours at high frequencies reconsidered

To add to the database and to clarify the spacing of the equal‐loudness contours at high frequencies, equal‐loudness relations determined from 1 to 16 kHz in a recent study [Hellman et al., J. Acoust. Soc. Am. 103, 2812 (1998)] are reevaluated. Relative to the linear loudness‐level function with a slope of 1.0 observed for a standard 1‐kHz tone, above 10 kHz the overall shapes and slopes of the loudness‐level functions are both level and frequency dependent. Below 60 phons, the slopes of the loudness‐level functions at 12.5 kHz and higher increase progressively with frequency from 1.31 at 12.5 kHz to 1.79 at 16 kHz. Conversely, above 60 phons the slopes decrease from 0.98 at 12.5 kHz to 0.74 at 16 kHz. The data imply that for frequencies between 1 and 10 kHz, the spacing between the equal‐loudness contours is independent of loudness level. In contrast, above 10 kHz the equal‐loudness contours are more closely spaced below 60 phons than at higher loudness levels. Moreover, in accord with the early work of ...