JP2000022469A - Audio processing unit - Google Patents

Abstract

PROBLEM TO BE SOLVED: To improve a response characteristics when a level of an input audio signal changes from a large one to a small one. SOLUTION: The processing unit is provided with a gain control amplifier 52 that receives an input audio signal and a level detection circuit 61 that detects a level of the input audio signal, and also a control signal generating circuit 63 that generates a control signal of a gain of the gain control amplifier 52 according to a detection signal of the level detection circuit 61. The gain control amplifier 52 has an input output characteristics where the received input audio signal is compressed to a level and the compressed signal is outputted. A response time of the gain control amplifier 52 whose gain changes from a small one to a large on by decreasing a large level of the input audio signal into a smaller level is changed depending on setting contents of level compression and a history of a level change in the input audio signal.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a speech processing apparatus used in the fields of hearing aids, telephones, loudspeakers, voice communications, and the like.

[0002]

2. Description of the Related Art When the dynamic range of an audio signal transmission system is narrow, or when the dynamic range that can be heard is narrow, such as a hearing-impaired person, for example, the straight lines 11 and 1 in FIG.
There is known a method of compressing the level of a target audio signal by an amplifier having input / output characteristics as shown in 2 or 13.

That is, these input / output characteristics 11 to 13
Are as shown by the straight lines 21 to 23. Regardless of the characteristic, the target audio signal is amplified with a large gain when the level is low, and is amplified with a small gain when the level is high. Is done. In the input / output characteristics 12 and the gain characteristics 22, the change range of the output level is narrower than in the input / output characteristics 11 and the gain characteristic 21, so that the level is more strongly compressed.
In the input / output characteristic 13 and the gain characteristic 23, the level compression ratio is the same as in the input / output characteristic 11 and the gain characteristic 21, but the output level and the gain are uniformly large.

[0004] In this way, the target audio signal is amplified and level-compressed at the same time, so that the dynamic range of the audio signal can be appropriately set within the dynamic range of the transmission system or the audible dynamic range.

[0005]

In the case of performing the level compression as described above, when the level of the audio signal changes, the time from the time of the change to the time when the gain of the amplifier changes to the corresponding gain (hereinafter, referred to as the level). If the "release time" is shortened, level compression is performed on the instantaneous level of the audio signal, so that the audio signal becomes a waveform that has undergone amplitude modulation and is greatly distorted, making it difficult to hear. turn into.

For this reason, in many cases, (A) when the audio signal changes from a small level to a large level, the gain is instantaneously reduced (the release time is shortened). (B) When the audio signal changes from a high level to a low level, the gain is increased with a predetermined time delay (the release time is lengthened). Like that.

However, when the release time is lengthened, (1) when a relatively loud sound enters temporarily while amplifying a small sound signal collected in a quiet environment, Even if the original sound becomes small, the gain remains at the small gain corresponding to the large sound until the release time ends, so that the small sound cannot be heard during that time.

[0008] (2) In an environment where the background noise level is high, when a telephone reception level is set to be higher than the background noise level and the user is listening to the sound signal, a temporary large sound signal is received.
Until the cancellation time ends, the gain remains small, and during that time, the small received voice is erased by the background noise and cannot be heard.

The present invention seeks to solve these problems.

[0010]

Therefore, according to the present invention, a gain control amplifier to which an input audio signal is supplied is provided.
A level detection circuit that detects the level of the input audio signal, and a control signal formation circuit that forms a control signal of the gain of the gain control amplifier according to the detection signal of the level detection circuit; The input audio signal supplied thereto has an input / output characteristic of level compression and output. The input audio signal decreases from a large level to a smaller level, so that the gain of the gain control amplifier corresponds to the input audio signal. According to another aspect of the present invention, there is provided an audio processing apparatus configured to change a response time from a small gain to a large gain according to the level compression setting and the history of the level change of the input audio signal. Therefore, the required gain is obtained even for a small input level after a large input level.

[0011]

In FIG. 1, an audio signal S51 to be processed is supplied to a gain control amplifier 52 through an input terminal 51, and is subjected to level compression in the control amplifier 52 before being taken out to an output terminal 53.

In this case, the control amplifier 52 can arbitrarily change the rate of change of the gain with respect to the input level, for example, as shown by the straight lines 24 and 25 in FIG. 2A according to the setting of the user. Yes, so that the dynamic range can be arbitrarily changed, as shown by the straight lines 14,15.

Further, the control amplifier 52 is configured such that the gain can be uniformly changed regardless of the input level, for example, as shown by the straight lines 26 and 27 in FIG. 3A according to the setting of the user. Therefore, as shown by the straight lines 16 and 17, the output level can be arbitrarily changed without changing the dynamic range.

A part of the audio signal S51 at the terminal 51 is
For example, a rectifier circuit 611 and a low-pass filter 612
, And the signal S
A detection signal (DC voltage) S61 whose level changes in accordance with the instantaneous level of the signal S51 is extracted. The signal S61 is supplied to the control signal forming circuit 63 as it is, and is also supplied to the delay circuit 62 to delay the signal S62 by a predetermined time.
The delay signal S62 is supplied to the forming circuit 63.

In this case, when there is a level change in the audio signal S51 from the difference between the signals S61 and S62, the amount of the change can be known, and the level of the signal S51 increases from the polarity of the difference between the signals S61 and S62. You can tell if it has become smaller or smaller. Therefore, in the forming circuit 63, the signal S6
1. A control signal S63 is formed from S62, and the gain of the gain control amplifier 52 is controlled by this signal S63, for example, as shown in FIGS. 2C and 3C.

That is, as shown in FIG. 2B, when the decrease in the input level is small, and when the dynamic range is set so as not to be too small, as shown by the straight line 14 (FIG. 2A), FIG. To curve 3
4, the gain changes relatively slowly, and as shown by the straight line 15 (FIG. 2A), when the dynamic range is set to be relatively small,
The gain is controlled to change relatively quickly, as shown by curve 35 in FIG. 2C.

In addition, as shown in FIG. 3B, when the input level is greatly reduced, and when the gain is uniformly set to a relatively small value as shown by a straight line 26 (FIG. 3A), FIG. The gain changes relatively slowly, as shown by curve 36 in FIG. 3, and when the gain is uniformly set relatively high, as shown by straight line 27 (FIG. 3A), as shown by curve 37 in FIG. In addition, the gain is controlled to change relatively quickly.

In an actual apparatus, the circuit shown in FIG. 1 is constituted by a digital circuit, and each signal is a digital signal. The main circuit is constituted by a DSP.

In such a configuration, it is assumed that the gain control amplifier 52 is set to have input / output characteristics as shown by a straight line 18 in FIG. 2A, for example.

Then, when the level of the input audio signal S51 changes from small to large to medium as shown in FIG. 4A, the gain of the control amplifier 52 becomes the input level as shown in FIG. 4B. When it changes from large to medium, it grows relatively quickly and can therefore respond quickly to the first small level.

However, when the level of the input audio signal S51 changes from middle to large to middle as shown in FIG. 5A, the gain of the control amplifier 52 becomes the input level as shown in FIG. 5B. When it changes from large to medium, it increases relatively slowly.

When the level of the input audio signal S51 changes from medium to large to small as shown in FIG. 6A, the gain of the control amplifier 52 becomes the input level as shown in FIG. 6B. Changes from large to small, it grows relatively quickly, and therefore can quickly respond to a small input level after a large input level.

Thus, according to this device, when recovering the gain after the input level is increased and the gain is reduced, the input / output characteristics and the history of the input level are taken into account, so that the large input level is obtained. The required gain even for small input levels after
It becomes easier to hear even small-level audio that follows a large level.

Further, since the level compression amplification is not instantaneously released, the amplitude modulation by the level of the audio signal is not received, and therefore, the audio with little distortion can be heard.

FIGS. 7 and 8 show a case where a bending point, that is, a knee point, is provided for the input / output characteristic of the gain control amplifier 52 in order to limit the input level range in which the amplification operation is performed.

In the case of FIG. 7, the characteristics 18, 19
In the input level above the knee point, the slope of the level compression is equal, but the knee point of the characteristic 19 is set lower than that of the characteristic 18. For this reason, the input level is large →
The rise of the gain when changing to a small value is made faster in the case of the characteristic 19 than in the case of the characteristic 18.

In the case of FIG. 8, the characteristics 18, 19
Has knee points set with different levels of level compression gradient. Therefore, when the input level changes from large to small, the gain rises faster in the case of the characteristic 19 than in the case of the characteristic 18.

FIG. 9 shows a case where the input audio signal S51 is divided into three frequency bands, and the same processing as that of the apparatus shown in FIG. 1 is performed in each band. That is, the audio signal S51 from the input terminal 51 is supplied to a band-pass filter 54L having a pass band of about 1 kHz or less, and the pitch of the signal S51 and the audio signal S51L of the first formant component are extracted from the filter 54L. .

The audio signal S51 from the terminal 51 is supplied to a band-pass filter 54M having a pass band of approximately 1 kHz to 3 kHz.
The audio signal S51M of the second formant component is extracted. Further, the audio signal S51 from the terminal 51 is almost 3
The signal is supplied to a high-pass filter 54H having a pass band equal to or higher than kHz, from which a third formant of the signal S51 and an audio signal S51H of a consonant component are extracted.

The signals S51L to S51H from the filters 54L to 54H are supplied to audio processing circuits 51L to 51H, respectively, which are configured similarly to the audio processing circuit shown in FIG. 1, and output from the gain control amplifiers 52 to 52. The supplied audio signals S51L to 51H are supplied to an addition circuit 55, and the added signals are taken out to an output terminal 52.

Accordingly, the level-compressed audio signal S51 is output to the terminal 55. According to the audio signal S51, even a low-level sound following a large level can be easily heard and the distortion is reduced. You can get less voice.

In the apparatus shown in FIG. 9, the audio signal S51
Can be divided, for example, by 1/3 octave or by a critical band.

[0033]

According to the present invention, the input / output characteristics and the history of the input level are taken into account when recovering the gain after the input level is increased and the gain amplification factor is reduced. It becomes easier to hear even small-level sound that follows. Also, since the level compression amplification is not instantaneously released, it is possible to hear a sound with little distortion.

[Brief description of the drawings]

FIG. 1 is a system diagram illustrating one embodiment of the present invention.

FIG. 2 is a diagram for explaining characteristics of the present invention.

FIG. 3 is a diagram for explaining characteristics of the present invention.

FIG. 4 is a diagram for explaining characteristics of the present invention.

FIG. 5 is a diagram for explaining characteristics of the present invention.

FIG. 6 is a diagram for explaining characteristics of the present invention.

FIG. 7 is a diagram for explaining characteristics of the present invention.

FIG. 8 is a diagram illustrating characteristics of the present invention.

FIG. 9 is a system diagram showing another embodiment of the present invention.

FIG. 10 is a diagram for explaining the present invention.

[Explanation of symbols]

51 = input terminal, 52 = gain control amplifier, 53 = output terminal, 61 = level detection circuit, 62 = delay circuit, 63 = control signal forming circuit

 ────────────────────────────────────────────────── ─── Continuing on the front page (72) Inventor Yoshihiko Deoka 6-7-35 Kita-Shinagawa, Shinagawa-ku, Tokyo F-term in Sony Corporation (reference) 5J029 AA01 BA05 CA02 CA04 EA01 5J030 BA06 BB00 BC02 BC07 BC08 5K027 BB03 DD14 DD16 DD18

Claims (7)

[Claims] A gain control amplifier to which an input audio signal is supplied; a level detection circuit for detecting a level of the input audio signal; and a control signal for controlling a gain of the gain control amplifier in accordance with the detection signal of the level detection circuit. The gain control amplifier has an input / output characteristic of level-compressing the input audio signal supplied thereto and outputting the input audio signal. By decreasing the gain to a small level, the response time until the gain of the gain control amplifier rises from the corresponding small gain to the large gain depends on the setting contents of the level compression and the history of the level change of the input audio signal. An audio processing device that can be changed. 2. The audio processing device according to claim 1, wherein the response time is shortened as the level compression ratio increases. 3. The audio processing device according to claim 1, wherein the response time is shortened as the gain setting of the gain control amplifier increases. 4. The response time of the gain change when the input level of the input audio signal changes from a first level to a second level which is larger than the first level. , The shorter the first level is, the shorter it is. 5. The response time of the gain change when the input level of the input audio signal changes from a first level to a second level which is larger than the first level. The sound processing device is characterized in that, when the level becomes smaller after the second level, the smaller the level, the shorter the level becomes. 6. The audio processing apparatus according to claim 1, wherein when the input level of the input audio signal is lower than a predetermined level, a lower limit is set to the input level so that the level compression operation is not performed. And the response time becomes shorter as the set lower limit becomes smaller. 7. The audio processing apparatus according to claim 1, wherein the input audio signal is divided into a plurality of frequency bands, and the divided audio signals in each frequency band are input to the input audio signal. An audio processing device.