Electrical audio signal processing systems and devices – Including frequency control – Automatic tone control
Reissue Patent
1999-05-10
2001-06-12
Isen, Forester W. (Department: 2747)
Electrical audio signal processing systems and devices
Including frequency control
Automatic tone control
C381S109000, C333S02800T
Reissue Patent
active
RE037223
ABSTRACT:
The present invention relates in general to equalizing and more particularly concerns novel apparatus and techniques for automatically controlling frequency response of a sound amplifying system as a function of level without introducing undesired boominess in voice reproduction. Conventional loudness contours for loudness compensation are described in a paper entitled “Loudness Compensation, Use and Abuse” by Tomlinson Holman & Frank S. Kapmann, Preprint No. 1281(D-4) of a paper presented at the Audio Engineering Society in November 1977. These loudness contour circuits are based upon Fletcher-Munson curves which relate the frequency response of a human to the level of the sound being heard. Humans are less sensitive to bass frequencies relative to other frequencies at lower volume levels than at higher levels. To compensate for this difference many audio amplifying systems include a loudness control that increases the bass and treble boost at lower volume levels.
The above-identified paper and U.S. Pat. No. 4,220,817 describe a bass-control circuit that attempts to match the shape of the Fletcher-Munson curves or those of later researchers. Historically, loudness controls in sound reproduction equipment have been designed so that as the volume control is adjusted, compensating networks change the frequency response of the system in the opposite manner to the way the ear changes its frequency response as the sound level is varied. It has been discovered that this approach leads to some undesirable effects. The present invention yields significantly improved performance.
To understand the problems with the existing loudness controls and how the present invention solves these problems, consider first how the frequency response of the human ear changes with the intensity of the sound presented to it. Then the effects of these changes on voice and music signals will be examined.
Equal-loudness contour curves were measured on subjects by Fletcher and Munson in the U.S.A. and later by Churcher and King in England. These equal-loudness contours show that as the sound level of pure tones presented to the ear is decreased, the ear loses its sensitivity to tones of frequencies below 500 Hz. Thus if an acoustic signal consisting of two pure tones of equal sound pressure level, one at 50 Hz and the other at 1,000 Hz, is experienced first at a high sound pressure level and then at a low pressure sound level, the signal of lower sound pressure level will be perceived to have less of the 50 Hz tone relative to the 1,000 Hz tone than will the signal of higher sound pressure level.
“Loudness” controls in sound reproduction equipment were developed to boost the low frequencies relative to the mid frequencies as the volume control is decreased in order to just compensate for the decreasing sensitivity of the ear to low frequencies as the sound level is decreased. However, boosting the low frequencies in signals reproduced at low sound levels does not restore the perception of the sound to that which could be experienced in a live situation. This is because the live sound, if heard at a low level, would also be subject to the ear's decreased sensitivity at low frequencies and would therefore be perceived to contain less low frequencies relative to the mid frequencies. For this reason loudness controls found on sound reproduction equipment today cause voices to be perceived as vary bass heavy (boomy) when reproduced at low volume levels. This is the reason that hi-fi equipment contains a switch to defeat the loudness control when it is objectionable. Daily, people have occasion to hear the live speech of others at different sound levels. This occurs, for example, with varying distances between the speaker and listener outdoors. It also occurs when people speak with different sound levels at different times. The lower sound levels of live speech are perceived to have less bass, and that is considered natural. It has been discovered that any attempt to restore these low frequencies to the voice in reproducing it at low sound levels is perceived as artificial.
However, experience with live music is generally quite different. Conductors do not play given compositions at, for example, 30 db different sound levels, and sound levels in concert halls do not differ by very large amounts in different seats. Thus in concert hall live performances listeners are accustomed to hearing a given composition at more or less a given sound level. Even with outdoor bands, concentrated listening is associated with a loud level in the immediate proximity of the band. In band music played outdoors without reinforcement one can of course move far enough away from the band to appreciably lower the sound level. However, as one does this, the air attenuates the high frequencies much more than the low frequencies, thus compensating in part for what would otherwise be a perceived loss of low frequencies with large distances.
When recorded music is played back at a sound level lower than that which would be experienced for that composition in a live performance, it is perceived that the bass instrument sounds (for example the double bass, bass drum and organ pedal notes) are virtually missing from the composition because of the decreased sensitivity of the ear to low frequencies. Since this low level listening experience is not (unlike the voice) associated with live performances, it was discovered that properly boosting the very low frequencies (below 200 Hz) restores perception of the bass instruments in a manner that is regarded as an improvement without degrading voice reproduction as do the prior art loudness compensation methods.
It has been discovered that even though the Fletcher-Munson equal-loudness curves predict one should also boost the frequencies between 200 and 500 Hz, very satisfactory musical performance is obtained by limiting the frequencies to be boosted to those below 200 Hz. In this manner the invention avoids the undesirable effects described earlier on the voice because the invention inserts negligible boost at any of the formant frequencies of the voice.
It has also been discovered that even below 200 Hz the optimum boost in frequency response is not, as is commonly believed, predicted by variations in the equal-loudness curves as a function of signal level. The Fletcher-Munson curves were made with pure tones (sine waves). They tell only what sound pressure level a pure tone of some given frequency must have in order to be perceived to be as loud as another pure tone at 1000 Hz of a given sound pressure level.
In music there is no fixed relationship between the amplitude of the low frequency components and those of the middle frequencies (which are largely responsible for establishing the level of loudness perceived from the musical composition) and therefore the volume control setting when reproducing music. The relative amplitudes of different tones depend upon the particular musical composition and vary within any composition.
Furthermore in music it is possible, for example, to have two different instruments, say an organ and a double bass, playing simultaneously in the same low frequency band with different amplitudes. Following the equal-loudness curve, theory of loudness compensation with volume control setting, a different amount of bass boost would be required for each of the instruments for a given decrease in the volume control setting. This, of course, cannot be achieved since the two music signal occupy the same frequency band and cannot be mathematically separated. Thus, it has been discovered that the equal-loudness contour curves are not the proper curves to use in the design of a loudness control.
According to the invention, there has been discovered a family of frequency contours which vary with volume control setting that improve the perceived low level music reproduction and do not degrade voice reproduction.
Accordingly, it is an important object of this invention to provide improved loudness compensation.
It is another object of the invention to provide improved automatic
Bose Amar G.
Plourde Richard G.
Bose Corporation
Fish & Richardson P.C.
Isen Forester W.
Mei Xu
LandOfFree
Dynamic equalizing does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Dynamic equalizing, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Dynamic equalizing will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-2458566