摘要:
A monolithic second order gradient (SOG) microphone structure employs acoustic transmission lines wherein the acoustic phase delay along each of the acoustic transmission lines is in direct proportion to the length of each of the acoustic transmission lines and, where this is effected by the use of an acoustic impedance element placed within each acoustic transmission line that has an acoustic impedance related to the acoustic impedance of the associated acoustic transmission line. In one embodiment, the acoustic impedance element has a specific acoustic impedance substantially matched to the specific acoustic characteristic resistance of the acoustic transmission line. Various embodiments may utilize acoustic or electrical subtraction of the signals in the acoustic transmission lines to realize the desired directional sound pickup.
摘要:
A "loudness balance" procedure is used to determine the amount of dynamic range for syllabic compression loss in an individual's cochlea. In particular, an individual is asked to adjust a set of tone volumes, which are played to them, until the tone volumes sound equally loud. Based on the adjustments performed by the individual, a compression ratio is determined for that individual over at least one frequency band. The determined compression ratio is then subsequently used by a telecommunications system to provide an individual-specific sound enhancement in a telephone call. As a result, the individual-specific sound enhancement particularly compensates for an amount of hearing loss unique to that individual.
摘要:
A new technique for the determination of the masking effect of an audio signal is employed to provide transparent compression of an audio signal at greatly reduced bit rates. The new technique employs the results of recent research into the psycho-physics of noise masking in the human auditory system. This research suggests that noise masking is a function of the uncertainty in loudness as perceived by the brain. Measures of loudness uncertainty are employed to determine the degree to which audio signals are "tone-like" (or "noise-like"). The degree of tone-likeness, referred to as "tonality," is used to determine masking thresholds for use in the compression of audio signals. Tonality, computed in accordance with the present invention, is used in conventional and new arrangements to achieve compression of audio signals.
摘要:
A new technique for the determination of the masking effect of an audio signal is employed to provide transparent compression of an audio signal at greatly reduced bit rates. The new technique employs the results of recent research into the psycho-physics of noise masking in the human auditory system. This research suggests that noise masking is a function of the uncertainty in loudness as perceived by the brain. Measures of loudness uncertainty are employed to form noise masking thresholds for use in the compression of audio signals. These measures are employed in an illustrative subband, analysis-by-synthesis framework. In accordance with the illustrative embodiment, provisional encodings of the audio signal are performed to determine the encoding which achieves a loudness differential, between the original and coded audio signal, which is less than (but not too far below) the loudness uncertainty.