JP3311580B2 - Audio signal compression device and camera - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an audio signal compression device and a camera for a video camera, an electronic still camera, a telephone and the like, and more particularly to an audio signal compression having an automatic gain control function.

[0002]

2. Description of the Related Art Audio signal compression devices are, for example, ISO / IEC
11172-3: 1993 (E) (MPEG (Moving Picture Experts G
roup) 1 audio). As shown in FIG. 6, the audio signal compression apparatus divides the entire frequency band of an audio input signal input from an input terminal 607 into 32 equally-spaced frequency components (subbands).
And a scaling circuit 602 for calculating a scale factor of each band (sub-band), and an adaptive bit allocation circuit 6 for calculating quantization bit allocation information of each sub-band
03, a quantization circuit 604, and a bit stream generation circuit 605 for converting a header, quantization bit allocation information, a scale factor, and quantized samples into a bit stream format.

[0003] For band division coding, a polyphase filter bank (polyphase filter bank) or MDCT (modified discrete cosine transform) is used, and highly efficient data is utilized using psychoacoustic characteristics. Compression is achieved.

The automatic gain control device includes a variable gain amplifier and a comparator for comparing an input signal level with a reference level, as described in, for example, Japanese Patent Application Laid-Open No. 1-170226. If the input signal is larger than the reference voltage, it is determined that the gain is too large, and the gain of the variable gain amplifier is reduced. If the input signal is smaller than the reference voltage, the gain is determined to be small and the gain of the variable gain amplifier is increased. In this configuration, the level of the signal is controlled by changing the gain of the variable gain amplifier according to the magnitude of the input signal.

[0005]

According to the above prior art, in order to realize automatic gain control in the audio signal compression processing system, a new automatic gain control circuit is added to the input section of the audio signal compression device. And the circuit configuration becomes complicated.

An object of the present invention is to provide an audio signal compression apparatus and a camera capable of realizing automatic gain control by adding relatively simple means.

Another object of the present invention is to provide an audio signal compression device and a camera having an automatic gain control function.

[0008]

SUMMARY OF THE INVENTION The present invention provides automatic gain control without adding new external means for automatic gain control by including automatic gain control processing in a series of processing for audio signal compression. This realizes the audio signal compression processing including the above.

Specifically, a band dividing means for dividing the entire frequency band of the input signal into a plurality of equally-spaced frequency components (sub-bands), and a reproduction magnification of each sub-band from the output signal of the band dividing means. A scaler for calculating a scale factor, a quantizer for quantizing sample data, and a bitstream generator for organizing a header, quantized bit allocation information, a scale factor, and quantized samples into a bitstream format; In an audio signal compression apparatus for generating a bit stream by compressing an input audio signal in a frame unit composed of a predetermined number of samples, an automatic gain control unit for controlling a scale factor of each subband calculated by the scaling unit is provided. Performs automatic gain control processing in the process of audio signal compression processing. Characterized in that it was.

[0010]

An embodiment of the present invention will be described below with reference to the drawings.

FIG. 1 is a block diagram showing an embodiment of an audio signal compression apparatus according to the present invention. In the figure, 1
01 is a band dividing circuit that divides the entire frequency band of the input signal into 32 equally-spaced frequency components (sub-bands), and 102 calculates the magnification (scale factor) at the time of reproduction based on the maximum amplitude of the sample of each sub-band. Scaling circuit,
Reference numeral 103 denotes an adaptive bit allocation circuit that calculates quantized bit allocation (bit allocation) information for each subband. Reference numeral 104 denotes an automatic bit rate calculator that calculates and outputs a new scale factor based on the scale factor calculated by the scaling circuit 102. A gain control circuit; 105, a quantization circuit for compressing original sound sample data in accordance with the scale factor and bit allocation information; 106, a bit stream generation circuit;
07.

The automatic gain control circuit 104 calculates and outputs a new scale factor value based on the scale factor input from the scaling circuit 102, as described above. It has the function of passing by value and outputting.

Reference numeral 108 denotes a microphone signal input terminal, 1
09 is an external signal input terminal, 110 is a selection switch for switching between input signals from the microphone signal input terminal 108 and the external signal input terminal 109, and 111 is an input control for controlling the automatic gain control circuit 104 and the selection switch 110. Circuit.

The operation of this embodiment will be described below.

The input control circuit 111 includes a selection switch 11
0 to the microphone signal input terminal 108 or the external signal input terminal 109,
1 receives an audio signal from a microphone or an external circuit, and controls the function of the automatic gain control circuit 104 in conjunction with the switching control of the selection switch 110. In the function control for the automatic gain control circuit 104, when the audio signal is input from the microphone signal input terminal 108 and processed, the automatic gain control circuit 104 functions so as to calculate a new scale factor. When the audio signal is input and processed, the automatic gain control circuit 104 functions so that the scale factor calculated and output by the scaling circuit 102 is passed as it is.

The input audio signal is 32 KHz,
It is a PCM signal sampled at a sampling frequency of either 44.1 KHz or 48 KHz.

The band dividing circuit 101 includes a selection switch 11
The audio signal selected and input at 0 is divided into 32 equally spaced sub-bands of frequency width. The band dividing circuit 101 is provided with a polyphase filter bank which has high calculation efficiency and does not cause aliasing (aliasing noise generated when signal processing is performed without cutting a frequency component higher than 1/2 of the sampling frequency). use.

The audio signal compression device 107 has an AAU
Signal processing is performed in a minimum unit (one frame) that can be decoded into an audio signal called (Audio Access Unit). 1
The data structure of the frame includes a header, bit allocation (quantization bit allocation of each subband) information, a scale factor, and samples. The number of samples in one frame is constant, and processing is performed, for example, using data of 384 samples (12 samples in one subband) as a unit.

The scaling circuit 102 extracts the sample having the maximum absolute value from the 12 samples of each sub-band divided into 32 sub-bands by the band division circuit 101, and extracts the sample having 63 steps (from +6 dB to -118 dB). 2d
Scale factor table (ISO / B units)
IEC 11172-3: 1993 (E)), find the scale factor table number (index) for which maximum amplitude ≤ scale factor table value, and the value of the scale factor table. Output the value as a scale factor.

The adaptive bit allocation circuit 103 converts the input signal into a block length of 512 bits or 1024 bits by FFT (Fast Fourier Transform / Fast Fourier Transform).
I do. Then, a masking threshold (a limit at which sound can be detected) is calculated from the result of the FFT and the scale factor,
The number of quantization bits (bit allocation information) to be allocated to each subband is determined. However, it is also possible to reduce the number of adaptive bit allocation circuits by using predetermined bit allocation information.

The automatic gain control circuit 104 calculates and outputs a new scale factor in consideration of gain based on the scale factor calculated by the scaling circuit 102 in the microphone signal input state, and inputs the new scale factor in the external signal input state. The scale factor is passed as it is and output.

In the quantization circuit 105, each of the subbands divided by the band division circuit 101 according to the scale factor calculated by the scaling circuit 102 and the number of quantization bits of each subband calculated by the adaptive bit allocation circuit 103. Quantize the sample.

The bit stream generation circuit 106 includes a header, bit allocation information, an automatic gain control circuit 1
The scale factor calculated in step 04 and the quantization circuit 105
The samples quantized in step (1) are organized into a bit stream format for each frame and output to a recording medium.

Since the scale factor is a reproduction magnification of 12 samples in each subband in one frame, the present invention changes (controls) the scale factor to change the gain of reproduced audio data. ing. In other words, by organizing a scale factor that is larger than the scale factor used in the quantization circuit 105 into a bit stream format, it is possible to increase the audio reproduction magnification. This is the same as increasing the gain at the time of input in the conventional automatic gain control device. Conversely, by organizing the reduced scale factor into a bit stream format, it is the same as reducing the audio reproduction magnification.

Here, a specific embodiment of the automatic gain control circuit 104 will be described with reference to the flowchart shown in FIG. 2 and FIGS. This flowchart shows the automatic gain control processing when the microphone signal is being input.

In FIG. 2, in step S 201, the scale factor table number of each (32) subband of one frame is fetched from the scaling circuit 102. In step S202, as shown in FIG. 7, the scale factor table number 7 for each (32) subband of one frame
01 to 732 are accumulated (cumulative value S ₀ ). Step S2
In 03, as shown in FIG. 8, the cumulative value S ₁₀ that accumulated a similar scale factor table number 801-832 and a subsequent cumulative value S ₀ was accumulated scale factor table number 701-732 to the total cumulative (total cumulative Value S _K ). In step S204, it is checked whether or not the accumulated values for 10 frames from the current frame to 10 frames before are totaled. In step S205, a difference (ΔS) between the total total value S _K and the total (S _K−1 ) from 11 frames before to 20 frames before is calculated. In step S206, when ΔS is positive, the scale factor table numbers of the 32 subbands fetched from the scaling circuit 102 are increased according to the magnitude of ΔS. Conversely, when ΔS is negative, the scale factor table number is decreased according to the magnitude of ΔS. However, since the scale factor table is in steps of about 2 dB, the minimum gain control width is about 2 dB.

Next, another embodiment of the automatic gain control circuit 104 will be described with reference to the flowchart shown in FIG. 3 and FIGS. This flowchart shows the automatic gain control processing when the microphone signal is being input.

In FIG. 3, in step S301, the scale factor table number of each subband of one frame calculated by the scaling circuit 102 is fetched. In step S302, as shown in FIG. 7, the scale factor table number 7 for each (32) subband of one frame
01 to 732 are accumulated (cumulative value S ₀ ). Step S3
03, as shown in FIGS. 4A and 4B,
The total value S ₀ of the scale factor table number 401 in the current frame, the total value S ₁ in the frame 402 one frame before, the total value S ₂ in the frame 403 two frames before, and the total value S ₉ in the frame 410 nine frames before. , The total value S ₁₀ of the frame 411 10 frames before, and 10 to 10
The total cumulative value S _L-1 in the total frame 412 to frame the previous frame 411, a total cumulative value S _L in total frame 413 from the current frame 401 up to 9 frames before frame 410, first, one frame before frame The total value S ₀ in the latest frame 401 is added to the total value S _L-1 in the total frame 412 from the frame 402 to the frame 411 10 frames before, and the total value S ₁₀ in the frame 411 10 frames before is subtracted. to seek a new total cumulative value S _L. In step S304 increases or decreases the scale factor table number calculated by scaling circuit 102 in accordance with the total cumulative value S _L.

As described above, by realizing the automatic gain control processing by controlling the scale factor in the course of the audio signal compression processing, the complexity of the circuit configuration can be reduced.

FIG. 5 is a block diagram of an MPEG camera according to the present invention. 5, reference numeral 501 denotes an imaging (lens unit, aperture unit, CCD unit) system; 502, an external video signal terminal; 503, an imaging system 501 and an external video signal terminal 502;
504 is an external audio signal terminal, 505 is a microphone system, 507 is a video signal compression device (MPEG CODEC) for compressing a video signal, and 508 is the audio signal compression device for compressing an audio signal according to the present invention. (MPEG AUDIO) and 509 are recording media for recording compressed video signals and audio signals, and an MPEG camera 510 is constituted by a set of these recording media.

The MPEG camera 510 includes an image pickup system 50
When used to record a video signal generated by photoelectric conversion in step 1 and an audio signal generated by the microphone system 505,
The above-described automatic gain control means in the audio signal compression device 508 according to the present invention effectively functions to compress the audio signal.

The present invention relates to a CPU or DSP operating at high speed.
It is desirable to use a signal processing LSI such as a (Digital Signal Processor) or the like and to effectively use functional means realized by processing with software.

[0033]

According to the present invention, since the automatic gain control processing is included in a series of processing for audio signal compression, it is not necessary to add a new external circuit for automatic gain control. With such a configuration, it is possible to realize an audio signal compression device having an automatic gain control function.

[Brief description of the drawings]

FIG. 1 is a block diagram showing an embodiment of an audio signal compression device according to the present invention.

FIG. 2 is a flowchart showing an example of an automatic gain control process executed by an automatic gain control circuit in the audio signal compression device shown in FIG.

FIG. 3 is a flowchart showing another example of the automatic gain control process executed by the automatic gain control circuit in the audio signal compression device shown in FIG.

FIG. 4 is a diagram illustrating the automatic gain control processing method shown in FIG.

FIG. 5 is a block diagram of an MPEG camera according to the present invention.

FIG. 6 is a block diagram of a conventional audio signal compression device.

FIG. 7 is a diagram for explaining the automatic gain control processing method shown in FIGS. 2 and 3;

8 is a diagram illustrating the automatic gain control processing method shown in FIG.

[Explanation of symbols]

101: band division circuit, 102: scaling circuit, 1
03: adaptive bit allocation circuit, 104: automatic gain control circuit, 105: quantization circuit, 106: bit stream generation circuit, 107: audio signal compression device, 108: microphone signal input terminal, 109: external signal input terminal, 110 ... Selection switch, 111 ... Input control circuit.

────────────────────────────────────────────────── ─── Continuation of the front page (56) References JP-A-1-170226 (JP, A) JP-A-8-123488 (JP, A) (58) Fields investigated (Int. Cl. ⁷ , DB name) H04N 5/92-5/928 G10L 19 / 02,21 / 02

Claims (6)

(57) [Claims] 1. A band dividing means for dividing an entire frequency band of an input audio signal into a plurality of sub-bands which are frequency components at equal intervals, and a scale factor which is a reproduction magnification of each sub-band from an output signal of the band dividing means. , A quantization means for quantizing the sample data, an automatic gain control means for controlling a scale factor of each subband calculated by the scaling means, a header, quantization bit allocation information, An audio signal compression device, comprising: a bit factor generator configured to organize a scale factor controlled by an automatic gain controller and sample data quantized by the quantizer into a bit stream format. 2. The audio signal compression apparatus according to claim 1, wherein said automatic gain control means calculates a new scale factor based on a scale factor of each subband calculated by said scaling means. 3. The automatic gain control means according to claim 1, wherein said automatic gain control means calculates a new scale factor based on an accumulated value obtained by accumulating the scale factor of each subband calculated by said scaling means for each frame. Audio signal compression device. 4. The automatic gain control means according to claim 1, wherein said automatic gain control means accumulates the scale factor of each subband calculated by said scaling means for each frame, and accumulates the accumulated value for each frame for a plurality of frames. An audio signal compression device for calculating a new scale factor based on a cumulative value of a plurality of frames. 5. A selection switch according to claim 1, wherein said input signal is selectively inputted from a microphone input terminal or an external input terminal, and said automatic gain control is performed when an input signal is inputted from a microphone input terminal. An audio signal compression device, wherein the operation of the means is enabled, and the operation of the automatic gain control means is disabled when inputting from an external input terminal. 6. An audio signal compression device according to claim 1, an image pickup system for generating a video signal, a microphone for generating an audio signal, and a video signal compression device for compressing the video signal. Recording means for recording a video signal compressed by the video signal compression device and an audio signal compressed by the audio signal compression device.