CN101009950A - A continuous-processing blind separation device for the mixed audio - Google Patents

Abstract

The invention discloses a continuous processing mixed audio blind separation device, which consists of a casing, a built-in circuit board, and a speaker and a sound box connected together, and a mixed audio signal input socket and a separation output working state indicator light are arranged on the panel of the casing , Separate the output speaker socket, the system reset button, and electrically connect with corresponding points on the circuit of the built-in circuit board respectively. The present invention can separate two kinds of mixed audio signals input continuously, output their respective audio signals independently, play them in real time by their respective speakers, and demonstrate and verify the practical application of the theory of signal blind separation realized by hardware circuits.

Description

A continuous processing hybrid audio blind separation device

Technical field

The invention relates to the technical field of signal blind separation, in particular to the application of mixed audio signal blind separation processing technology, in particular to a continuous processing type mixed audio blind separation device.

Background technique

The theory of blind signal separation is one of the popular research topics in the field of signal processing today. It refers to the use of mixed signals generated by the source only in the so-called "blind" state without knowing the specific source and channel. are blind signals), and through corresponding processing, the respective source signals can be separated (called blind separation). This theory has a very broad application prospect. However, most of the application research and effect demonstration of the blind signal separation theory are still in the stage of computer software simulation, considering the ideal test conditions and environment, and due to the limitation of the processing speed in the machine, the mixed signal input of the demonstration can only be divided into three parts. Segment and intermittent, there is no off-line demonstration and real-time application device realized by hardware circuit.

Contents of Invention

The object of the present invention is to provide a continuous processing blind separation device for mixed audio in order to solve the above-mentioned shortcomings in the prior art. The device realizes blind signal separation that can be demonstrated offline and processed in real time through hardware circuits, and can continuously separate two mixed audio signals, output their own audio signals independently, and play them in real time by their respective speakers.

The purpose of the present invention is achieved through the following technical solutions: the continuous processing type mixed audio blind separation device comprises a casing connected with a built-in circuit board and a speaker box, and the panel of the casing is provided with a mixed audio signal The input socket, the separation output working state indicator light, the separation output horn socket, the system reset button are electrically connected to the corresponding points of the circuit on the built-in circuit board respectively.

In order to better realize the present invention, the circuit on the built-in circuit board includes a channel 1 mixed input audio amplifier circuit, a channel 1 serial encoding/decoding/filtering circuit, a channel 1 negative voltage generating circuit, and a channel 1 separated output audio power amplifier Circuit, channel 2 mixed input audio amplifier circuit, channel 2 serial encoding/decoding/filtering circuit, channel 2 negative voltage generating circuit, channel 2 separated output audio power amplifier circuit, digital signal processing circuit, working status indication circuit, system reset circuit 1. The system simulation debugging interface and the power supply voltage type conversion circuit are jointly electrically connected, and the interconnection relationship is as follows: the channel 1 serial encoding/decoding/filtering circuit passes the channel 1 mixed input audio amplifier circuit signal line, and the channel 1 negative voltage generates The signal line of the circuit, the signal line of the channel 1 separated output audio power amplifier circuit are electrically connected with the mixed input audio amplifier circuit of channel 1, the negative voltage generating circuit of channel 1, and the separated output audio power amplifier circuit of channel 1, and the channel 2 is serially coded The /decoding/filtering circuit mixes the input audio amplifier circuit signal line of channel 2, the negative voltage generation circuit signal line of channel 2, and the separated output audio power amplification circuit signal line of channel 2 with the mixed input audio amplifier circuit of channel 2 and the negative voltage generation circuit of channel 2 respectively. The circuit and the channel 2 separate output audio power amplifier circuit are electrically connected, and the digital signal processing circuit is connected through the channel 1 serial encoding/decoding/filtering circuit signal line, the channel 2 serial encoding/decoding/filtering circuit signal line, and the working status indication Circuit signal line, system reset circuit signal line, system simulation debugging interface signal line, power supply voltage type conversion circuit signal line and channel 1 serial encoding/decoding/filtering circuit, channel 2 serial encoding/decoding/filtering circuit, working status The indication circuit, the system reset circuit, the system simulation debugging interface and the power supply voltage type conversion circuit are electrically connected.

Compared with the prior art, the present invention has the following advantages and beneficial effects:

1. The present invention realizes blind separation processing of mixed audio signals through hardware circuits.

2. The present invention can separate the continuous input of two kinds of mixed audio signals in real time, and perform playback and demonstration by respective speakers.

3. The present invention is compact and portable, so that the application demonstration of the blind signal separation theory can be separated from the level of computer equipment and its software simulation.

Description of drawings

Fig. 1 is a structural representation of the present invention;

Fig. 2 is a circuit block diagram of the present invention;

Fig. 3 is a schematic circuit diagram of the present invention.

Detailed ways

The present invention will be described in further detail below with reference to the drawings and embodiments, but the embodiments of the present invention are not limited thereto.

As shown in Figure 1, the panel of the casing 1 is provided with channel 1 mixed audio signal input socket 2, channel 2 mixed audio signal input socket 3, channel 2 separated output working status indicator light 4, channel 1 separated output working status indicator Light 5, system reset button 6, channel 1 separate output speaker socket 7, channel 2 separate output speaker socket 8, DC power supply lead-in wire 13, lead-in wires 9 and 10 of speaker speakers 11 and 12 are respectively connected to the separate output speaker socket on the panel 7 and 8 are connected.

As shown in Figure 2, the working principle of the present invention is: two kinds of mixed audio analog signals that need to be separated are respectively sent to channel 1 mixed input audio amplifier circuit 14 and channel 2 mixed input audio amplifier circuit 18 for amplifying processing, after amplifying The final output signals are respectively sent to the respective channel 1 serial encoding/decoding/filtering circuit 15 and channel 2 serial encoding/decoding/filtering circuit 19 for A-D (analog-digital) conversion and filtering, and then the obtained mixed digital The signal is sent to the digital signal processing circuit 22 for audio separation processing; the channel 1 negative voltage generating circuit 16 and the channel 2 negative voltage generating circuit 20 respectively provide the channel 1 serial encoding/decoding/filtering circuit 15 and the channel 2 serial encoding/decoding/ The negative power supply voltage of the filter circuit 19; the digital signal processing circuit 22 returns the separated two digital signals to the channel 1 serial encoding/decoding/filtering circuit 15 and the channel 2 serial encoding/decoding/filtering circuit 19 for processing D-A (digital-to-analog) conversion processing, then the independent audio analog signals obtained are respectively output to the respective channel 1 separation output audio power amplifying circuit 17 and channel 2 separation output audio power amplifying circuit 21, and are played out from respective loudspeakers; At the same time, the digital signal processing circuit 22 displays the working states of the channel 1 serial encoding/decoding/filtering circuit 15 and the channel 2 serial encoding/decoding/filtering circuit 19 through the indicator light of the working state indicating circuit 23; the system reset circuit 24 can Make the working state of the digital signal processing circuit 22 start from scratch; in the system debugging stage, the external computer can realize program input, debugging and control to the digital signal processing circuit 22 through the system simulation debugging interface 25; the power supply voltage type conversion circuit 26 provides digital signal processing Various operating voltages required by the circuit 22.

As shown in Figure 3, the channel 1 mixed input audio amplifier circuit consists of an integrated part U1, a mixed signal input socket J1, resistors R1~R7, and capacitors C1~C5 are electrically connected together; the channel 1 serial encoding/decoding/filtering circuit includes an integrated U3, resistors R8, R9, capacitors C6, C7 are electrically connected together; channel 1 negative voltage generating circuit includes integrated parts U5, capacitors C8, C9 are electrically connected together; channel 1 separate output audio power amplifier circuit includes integrated parts U7, Resistors R-1, R10, capacitors C10~C14, and speaker SP1 are electrically connected together; channel 2 mixed input audio amplifier circuit includes integrated part U2, mixed signal input socket J2, resistors R11~R17, capacitors C15~C19 are electrically connected together ; The channel 2 serial encoding/decoding/filtering circuit consists of an integrated part U4, resistors R18, R19, capacitors C20 and C21, which are electrically connected together; the channel 2 negative voltage generating circuit consists of an integrated part U6, capacitors C22 and C23, which are electrically connected together; Channel 2 separate output audio power amplifier circuit consists of integrated parts U8, resistors R-2, R20, capacitors C24~C28, and speaker SP2 are electrically connected together; digital signal processing circuit includes integrated parts U9, crystal oscillator CY, capacitors C29, C30 Composed of common electrical connections; the working state indication circuit consists of light-emitting diodes LED1, LED2, resistors R21, R22, and common electrical connections; the system reset circuit consists of key switches SW, resistors R23, and capacitors C31; the system simulation debugging interface includes standard formats JTAG socket J3 is electrically connected; the power supply voltage type conversion circuit includes three-terminal voltage stabilizing integrated parts U10, U11, and capacitors C32-C38 are electrically connected together.

The working principle of the circuit of the present invention is: the two mixed audio signals to be separated enter from J1 and J2 respectively at the same time, are blocked by C1 and C15, filtered by R2, C3 and R12, and C17, and then sent to the operational amplifiers U1 and U2 Negative input terminal, the positive input terminal reference voltage of U1 and U2 is provided by R3, R4, R13 and R14, and the feedback loop of R6, C4 and R16, C18 determines the amplification of U1 and U2, so that the signals of U1 and U2 The amplification range is consistent; the amplified mixed analog signal is respectively DC-blocked by C5 and C19, and current-limited by R7 and R17, and sent to the positive serial input terminals VFXI+ of the respective serial encoding/decoding/filtering integrated parts U3 and U4, And R9, R8 and R19, R18 are used as the feedback loop of the negative input terminal VFXI-; the clock signals of the serial clock input terminals BCLKX and MCLKX come from the CLKX1, CLKR1 and CLKX0, CLKR0 terminals of the digital signal processing integrated part U9 respectively, and The sending frame synchronization terminals FSX1 and FSX0 of the digital signal processing integrated part U9 respectively provide sending frame synchronization signals to the FSR terminals of U1 and U2, and the receiving frame synchronization terminals FSR1 and FSR0 of U9 provide receiving frame synchronization signals to the FSX terminals of U1 and U2 respectively. signal to realize U9’s signal sending and receiving control to U3 and U4; the serial data sent by U3 and U4 to U9 is output from the serial data sending end DX and sent to the serial data receiving end DR1 and DR0 of U9 respectively, while U3, U4 The serial data received by the serial data receiving end DR of U4 from U9 is sent by the serial data sending ends DX1 and DX0 of U9 to realize the serial data transmission between U3, U4 and U9; the integrated parts U5 and U6 generate and The negative voltage corresponding to the positive voltage VCC is sent from the output terminal OUT to the negative voltage input terminal VBB of U3 and U4 respectively, providing the negative working voltage required by U3 and U4; the mixed audio signal sent from VFXI+ is respectively in U3 and U4 U4 internally performs A-D (analog-digital) signal conversion processing to become a mixed digital signal. Under the control of the clock signal of U9 and the serial synchronous control signal, it is sent from the serial data sending end DX to the serial data receiving terminal of U9 respectively. Terminals DR1 and DR0, separate the mixed digital signals in U9 to become two independent digital audio signals, and then under the control of the clock signal of U9 and the serial synchronous control signal, the signals from the serial data sending terminal DX1 and DX0 are looped back to the serial data receiving end DR of U3 and U4; the independent digital audio signals are processed by D-A (digital-analog) signal processing and corresponding filtering in U3 and U4 respectively, and become two independent audio analog signals , and then output from the analog signal output terminal VFRO respectively, and send to the respective audio power amplifier stages U7 and U8 for power amplification; the audio signal enters from the + input terminal of the power amplifier integrated parts U7 and U8 respectively, R-1, C10 The feedback loop with R-2 and C24 is used for amplification control, and R10, C11, C12, C14 and R20, C25, C26, and C28 are respectively used for signal filtering, coupled with audio signals by C13 and C27, and played through their respective speakers SP1 and SP2 At the same time, U9 controls the on-off of light-emitting diodes LED1 and LED2 through the control ports HD1 and HD2, so that they light up when U3 and U4 have separate audio signal outputs, and go out when there is no audio signal output, providing system Working status indication; when the button SW is pressed, a low level is provided to the reset terminal RESET of U9 to realize system reset; J3 is a standard format JTAG simulation debugging interface, which provides a system simulation debugging channel for the external computer and U9, and realizes the external computer to Program input, debugging and control before U9 runs off-line; three-terminal voltage regulator integrated parts U10 and U11 use DC power supply VCC to generate low voltage DVDD and CVDD required for U9 to work, and C32-C38 are used for power supply filtering.

The inventor thinks that the preferred mode of realizing the present invention can be: (1) by shown in Fig. 1, design and manufacture casing 1, for example, can select plastic sheet to be processed into 180 * 100 * 40mm with bonding method Case 1; On the panel of case 1, there are mixed audio signal input socket holes 2 and 3, separated output working status indicator holes 4 and 5, system reset button hole 6, and separated output speaker socket hole 7 And 8, DC power supply lead-in hole position and power lead-in 13 thereof; In addition, be equipped with loudspeaker box 11 and 12 and signal lead-in 9 and 10 thereof; (2) as one of embodiment of the present invention, as Fig. 2 and Fig. As shown in 3, make the built-in circuit board, and install and connect after screening the components. For example, the integrated parts U1 and U2 can be LM324, U3 and U4 can be TP3057, U5 and U6 can be MAX660, and U7 and U8 can be MAX660. Choose LM386 type, U9 can choose C6711 type, U10 and U11 can choose LT1585 type, crystal oscillator CY can choose 40MHZ type, speaker SP1 and SP2 can choose 8 ohm 2 watt type, the whole machine VCC uses 5V power supply, DVDD is 3.3 V, CVDD is 1.8V; the circuit board can be installed and connected according to the connection relationship shown in Figure 2 and Figure 3 and the above, and electrically connected with related components. After the circuit board is installed and connected, it is placed in the case. Carry out power-on debugging, just can implement the present invention preferably.

Claims (10)

1, a kind of continuous processing formula mixed audio blind separation device, it is characterized in that, comprise that casing and internal circuit board and Horn loudspeaker box connect and compose jointly, the panel of described casing is provided with the mixed audio signal input socket, separates the output services status indicator lamp, separates output loudspeaker socket, system reset pad, and respectively with internal circuit board on the circuit respective point be electrically connected mutually.

2, a kind of continuous processing formula mixed audio blind separation device according to claim 1, it is characterized in that, circuit on the described internal circuit board comprises that passage 1 mixes the input audio amplifier circuit, passage 1 serial code/decoding/filter circuit, passage 1 negative voltage generation circuit, passage 1 separates the output audio power amplification circuit, passage 2 mixes the input audio amplifier circuit, passage 2 serial code/decoding/filter circuits, passage 2 negative voltage generation circuit, passage 2 separates the output audio power amplification circuit, digital signal processing circuit, the operating state indicating circuit, system reset circuit, system emulation debugging interface and supply power voltage type conversion circuit are electrically connected formation jointly, its interconnected relationship is: described passage 1 serial code/decoding/filter circuit mixes input audio amplifier circuit holding wire by passage 1, passage 1 negative voltage generation circuit signal line, passage 1 separates output audio power amplification circuit holding wire and mixes the input audio amplifier circuit respectively with passage 1, passage 1 negative voltage generation circuit, passage 1 separates the output audio power amplification circuit and is electrically connected mutually, described passage 2 serial code/decoding/filter circuits mix input audio amplifier circuit holding wire by passage 2, passage 2 negative voltage generation circuit signal lines, passage 2 separates output audio power amplification circuit holding wire and mixes the input audio amplifier circuit respectively with passage 2, passage 2 negative voltage generation circuit, passage 2 separates the output audio power amplification circuit and is electrically connected mutually, and described digital signal processing circuit is by passage 1 serial code/decoding/filter circuit holding wire, passage 2 serial codes/decoding/filter circuit holding wire, operating state indicating circuit holding wire, the system reset circuit holding wire, system emulation debugging interface holding wire, supply power voltage type conversion circuit signal line respectively with passage 1 serial code/decoding/filter circuit, passage 2 serial code/decoding/filter circuits, the operating state indicating circuit, system reset circuit, the system emulation debugging interface is electrically connected mutually with supply power voltage type conversion circuit.

3, a kind of continuous processing formula mixed audio blind separation device according to claim 2, it is characterized in that described passage 1 mixes the input audio amplifier circuit and comprises that integration member U1, mixed signal input socket J1, resistance R 1～R7, capacitor C 1～C5 are electrically connected formation jointly.

4, a kind of continuous processing formula mixed audio blind separation device according to claim 2 is characterized in that described passage 1 serial code/decoding/filter circuit comprises that integration member U3, resistance R 8, R9, capacitor C 6, C7 are electrically connected formation jointly; Described passage 1 negative voltage generation circuit comprises that integration member U5, capacitor C 8, C9 are electrically connected formation jointly.

5, a kind of continuous processing formula mixed audio blind separation device according to claim 2, it is characterized in that described passage 1 separates the output audio power amplification circuit and comprises that integration member U7, resistance R-1, R10, capacitor C 10～C14, loudspeaker SP1 are electrically connected formation jointly.

6, a kind of continuous processing formula mixed audio blind separation device according to claim 2, it is characterized in that described passage 2 mixes the input audio amplifier circuit and comprises that integration member U2, mixed signal input socket J2, resistance R 11～R17, capacitor C 15～C19 are electrically connected formation jointly.

7, a kind of continuous processing formula mixed audio blind separation device according to claim 2 is characterized in that described passage 2 serial code/decoding/filter circuits comprise that integration member U4, resistance R 18, R19, capacitor C 20, C21 are electrically connected formation jointly; Described passage 2 negative voltage generation circuit comprise that integration member U6, capacitor C 22, C23 are electrically connected formation jointly.

8, a kind of continuous processing formula mixed audio blind separation device according to claim 2, it is characterized in that described passage 2 separates the output audio power amplification circuit and comprises that integration member U8, resistance R-2, R20, capacitor C 24～C28, loudspeaker SP2 are electrically connected formation jointly.

9, a kind of continuous processing formula mixed audio blind separation device according to claim 2 is characterized in that described digital signal processing circuit comprises that integration member U9, crystal oscillator CY, capacitor C 29, C30 are electrically connected formation jointly.

10, a kind of continuous processing formula mixed audio blind separation device according to claim 2 is characterized in that described operating state indicating circuit comprises that LED 1, LED2, resistance R 21, R22 are electrically connected formation jointly; Described system reset circuit comprises that key switch SW, resistance R 23, capacitor C 31 are electrically connected formation jointly; Described system emulation debugging interface comprises that reference format JTAG socket J3 is electrically connected formation; Described supply power voltage type conversion circuit comprises that three-terminal voltage-stabilizing integration member U10, U11, capacitor C 32～C38 are electrically connected formation jointly.

Images