Data processing: speech signal processing – linguistics – language – Speech signal processing – For storage or transmission
Reexamination Certificate
1999-12-30
2002-06-11
{haeck over (S)}mits, Tãlivaldis Ivars (Department: 2641)
Data processing: speech signal processing, linguistics, language
Speech signal processing
For storage or transmission
C704S500000, C381S023100
Reexamination Certificate
active
06405164
ABSTRACT:
FIELD OF THE INVENTION
The present invention relates generally to compressor circuits for audio signals. More particularly, the present invention relates to audio compressors for hearing aids.
BACKGROUND OF THE INVENTION
Audio compressors are devices that are commonly used to modify the dynamic range of an audio signal. Generally, an audio compressor comprises an amplifier and a control system to control the gain of the amplifier as a function of the input and/or output signal strength. An audio compressor circuit reduces the gain of an amplifier at high input signal levels to prevent distortion which would otherwise occur when the input signal exceeds a threshold level. However, audio compressors also provide other useful benefits.
Commonly, an audio compressor is characterized in terms of a compression ratio, which is the ratio of the gain at a reference signal level divided by the gain at a specified higher signal level. For example, an audio compressor with a 3:1 compression ratio has the characteristic that each 3 dB increase in sound level beyond a threshold level results in a 1 dB increase in output. However, more generally an audio compressor may be characterized in terms of a compression function that defines a relationship between input signal strength and the differential gain of the amplifier.
A variety of common sounds, such as the clang of a bell, have a sound amplitude with a high initial amplitude which decays over time. It is desirable that a compressor reduce gain rapidly enough that the amplification of such sounds does not cause unpleasant distortion and/or an excessive sound-level. However, a highly unnatural sound results if the gain recovers to its ordinary level too rapidly after an initial loud sound. Consequently, the transient response of a compressor has a large impact on the sound quality. Audio compressors are characterized in terms of an “attack” time, the time it takes for the amplifier to stabilize to within about 1 dB of its final increased value after a sudden increase in input audio level and a “release” time associated with the time it takes for an audio compressor to return to within about 1 dB of its final decreased value after the input audio level has decreased to normal levels. Typically, it is desirable that an audio compressor have a release time that is much longer than its attack time. Typical release times are often in the range of 1-10 milliseconds. Typical attack times are in the range of 0-1 milliseconds.
Audio compressors are commonly used in hearing aids. One reason that audio compressors are used in hearing aids is that hearing loss is often dependent upon both the sound frequency and the audio level. The ability to hear high frequency sounds may, for example, depend upon whether or not the listener is in a quiet or noisy environment. Typically less amplification is required if the background audio level is high. Thus, the required amplification will depend upon both the sound frequency and the audio level. An audio compressor that varies the amplification as a function of input audio level is thus useful for those types of hearing losses for which the input audio level determines the required amplification.
Audio compressors are also used in hearing aids to reduce distortion. The human ear is very sensitive to harmonic distortion. As is well known in the field of amplifier design, there is typically a preferred range of input signal levels for which low-noise linear amplification is possible for a particular Class-A, Class-AB, Class-B, or Class-D amplifier circuit. An output amplifier that is driven close to its maximum power output typically produces substantial harmonic distortion as the amplifier gain becomes saturated.
The design of an audio compressor for a miniature hearing aid is constrained by several design considerations that are unique to miniature hearing aids. First, a miniature hearing aid is typically powered by a single miniature hearing aid battery, which provides a nominal 1.5 VDC and which is capable of only limited load currents. High performance microprocessor circuits, by way of comparison, typically operate at a voltage of 3.3 VDC. Second, total circuit size is a major concern in the design of a hearing aid, since modem miniature hearing aids are designed to fit partially or totally in the ear canal. Third, cost is an issue. While high performance hearing aids sell for several thousand dollars, there is a world-wide mass market for hearing aids with a price of less than about two-hundred dollars. Moreover, since the total cost to a user includes audiologist fees, it is desirable that the characteristics of the hearing aid may be quickly adjusted to accommodate the user's needs and preferences.
Conventional audio compressors used in hearing aids commonly comprise an op-amp amplifier circuit whose gain is controlled by a current-controlled resistor.
FIG. 1
is a circuit schematic of a prior art hearing aid circuit. A microphone
2
detects input sounds. An output amplifier stage
4
is used to provide large-signal amplification to provide an amplified signal to a speaker/transducer
6
. A small signal amplifier and compression stage
8
includes an op-amp
16
with an input resistor
14
and a current-controlled resistor element
12
. The compression function of small signal amplifier and compression stage
8
is achieved by gain control circuit
10
adjusting the resistance of current controlled resistor element
12
as a function of the audio signal input level. One common gain control circuit
10
uses the logarithmic current-voltage output of the collector-emitter current of a bipolar transistor in response to increasing base voltage in order to control the input currents to current controlled resistor element
12
. However, there is only a limited current range over which a bipolar transistor has a current-voltage relationship that approximates a logarithmic function. Conventional current-controlled resistors also have numerous well-known limitations in regards to their accuracy and sensitivity.
There are several other drawbacks to the compressor circuit of
FIG. 1
, particularly when the audio compressor stage
8
, output stage
4
and other electronics (not shown in
FIG. 1
) are fabricated on one integrated circuit. Bipolar transistors require a substantial circuit area compared to complimentary metal oxide semiconductor (CMOS) transistors. A hearing aid circuit fabricated using an all bipolar transistor process occupies a substantial circuit area, which increases its cost. Moreover, bipolar transistors consume more current and power than CMOS transistor circuits. While a mixed circuit fabrication process combining both CMOS transistors and bipolar transistors on one chip (commonly called a “bi-CMOS” process) is technically feasible, a bi-CMOS fabrication process is typically expensive compared to fabricating either an all-bipolar or all-CMOS circuit.
The function of small signal amplifier and compression stage
8
may also be implemented using a digital signal processing circuit (DSP). DSP circuits permit complex signal processing functions to be performed, permitting an arbitrary gain compression function to be achieved. However, DSP circuits typically comprise substantial memory, logic, and control circuit elements. A DSP circuit will tend to consume a substantial chip area and may draw a significant amount of current compared to the hearing aid output amplifier stage
4
. Generally, utilizing a DSP circuit to perform an audio compression function for the output amplifier stage
4
of a hearing aid would substantially increase the size, cost, and complexity of a hearing aid circuit.
Unfortunately, conventional audio compressor circuits used in hearing aids tend to be expensive, requires a large area on an integrated circuit, and suffer from other problems, such as difficulties in providing control of audio compression over a wide range of audio input signal strengths. Moreover, most common audio compressor circuits provide only a limited ability to control and select the audio compress
Abebe Daniel
Coudert Brothers LLP
Engineering Consortium, Inc.
{haeck over (S)}mits Tälivaldis Ivars
LandOfFree
Audio compression circuit and method does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Audio compression circuit and method, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Audio compression circuit and method will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-2910124