WO2017071183A1 - Voice processing method and device, and pickup circuit - Google Patents

Abstract

A voice processing method and device, and a pickup circuit. The method comprises: collecting a voice signal, the voice signal comprising a voice incoming signal inputted by a user, and a voice feedback signal broadcast by a voice recognition processor (S11); eliminating the voice feedback signal in the voice signal to obtain the voice incoming signal (S12); and transmitting the voice incoming signal to the voice recognition processor which is used for identifying the voice incoming signal and broadcasting the voice feedback signal (S13). By means of simultaneously collecting the voice incoming signal inputted by the user and the voice feedback signal outputted by the voice recognition processor and eliminating the voice feedback signal therein, the final collected voice signal is not interfered by the feedback signal, allowing for a ''clean'' voice incoming signal to obtained, and thereby improving voice recognition accuracy.

Description

Voice processing method, device and sound collecting circuit

This application is based on an invention patent application filed on October 29, 2015, the application number is CN201510719799.4, entitled "A Voice Processing Method, Apparatus, and Sound Picking Circuit", and claims the priority of the invention patent application. The entire disclosure of this patent application is incorporated herein by reference.

Technical field

The present invention relates to the field of voice recognition technology, and in particular, to a voice processing method, apparatus, and sound pickup circuit.

Background technique

With the development of artificial intelligence technology, voice as a good human-computer interaction mode has gradually been applied to many intelligent electronic devices. Voice input is an input method for converting a person's spoken content into text by voice recognition. In many fields, the user can perform the corresponding command by means of voice input instead of manual input, and at the same time, the electronic device feeds back in the form of voice announcement. For example, if the user voice inputs "play the next song", the electronic device that receives the voice input recognizes the voice content input by the user, and switches the currently played song to the next song according to the voice command. It can be seen that the voice input can bring great convenience and fun to the user's life and work. However, in the process of voice input, when the speaker broadcasts the feedback front end through the voice broadcast, it will form a loop with the front end microphone, so that the user voice input When the voice signal received by the microphone is mixed with the interference of the feedback signal, the accuracy of the voice recognition is lowered.

Summary of the invention

The embodiment of the invention provides a voice processing method, a device and a sound collecting circuit, which are used for eliminating noise in a user's voice input signal, thereby improving the accuracy of voice recognition.

In a first aspect, a speech processing method is provided for use in a sound pickup circuit, the method comprising the steps of:

Acquiring a voice signal, the voice signal including a voice input signal input by the user and a voice feedback signal broadcast by the voice recognition processor;

Eliminating a voice feedback signal in the voice signal to obtain the voice input signal;

Transmitting the speech input signal to the speech recognition processor, the speech recognition processor for identifying the speech input signal and broadcasting the speech feedback signal.

Some beneficial effects of embodiments of the present invention may include:

The above technical solution, by simultaneously collecting the voice input signal input by the user and the language broadcasted by the voice recognition processor The sound feedback signal, and the speech feedback signal is eliminated, so that the finally collected speech signal is not interfered by the feedback signal, thereby obtaining a "clean" speech input signal, thereby improving the accuracy of speech recognition.

In one embodiment, the canceling the voice feedback signal in the voice signal to obtain the voice input signal includes:

The speech signal and the speech feedback signal are subtracted to obtain the speech input signal.

In this embodiment, by subtracting the voice signal and the voice feedback signal, a "clean" voice input signal can be obtained, so that the collected voice signal is not interfered by the feedback signal.

In one embodiment, the transmitting the voice input signal to the voice recognition processor comprises:

Performing analog-to-digital conversion on the voice input signal to obtain a digital voice input signal;

Performing a preset process on the digital voice input signal, the preset process including any one or more of a noise reduction process, a canceling reverberation process, and an echo cancellation process;

Transmitting the processed digital speech input signal to the speech recognition processor.

In this embodiment, by performing analog-to-digital conversion on the voice input signal, and processing such as noise reduction, reverberation elimination, and echo cancellation, the speech recognition signal received by the speech recognition processor has no noise interference, thereby enabling more accurate recognition. Voice input signal.

In one embodiment, the collecting a voice signal includes:

Receiving an acquisition instruction sent by the voice recognition processor, where the acquisition instruction includes a sampling clock of a reset signal and a voice signal;

Acquiring a voice signal according to the acquisition instruction.

In this embodiment, the acquisition of the voice signal is started according to the acquisition instruction sent by the voice recognition processor, and the reset control of the circuit is realized.

In an embodiment, before the acquiring the voice signal, the method further includes:

Identifying the voice signal;

Determining whether the voice signal is a valid voice signal according to a preset effective voice signal feature;

And when the voice signal is the valid voice signal, outputting an execution instruction to the voice recognition processor, the execution instruction is used to instruct the voice recognition processor to issue an acquisition instruction, where the collection instruction is used to indicate the The sound pickup circuit collects a voice signal.

In this embodiment, by recognizing the received speech signal and instructing the speech recognition processor to issue an acquisition instruction when the speech signal is a valid speech signal, the speech recognition processor wakes up the pickup circuit, thereby avoiding picking up The tone circuit receives an invalid voice signal.

In a second aspect, a speech processing apparatus is provided for use in a sound pickup circuit, the apparatus comprising:

An acquisition module, configured to collect a voice signal, where the voice signal includes a voice input signal input by a user and a voice feedback signal broadcast by the voice recognition processor;

a cancellation module, configured to cancel a voice feedback signal in the voice signal, to obtain the voice input signal;

And a transmission module, configured to transmit the voice input signal to the voice recognition processor, where the voice recognition processor is configured to identify the voice input signal and broadcast the voice feedback signal.

In one embodiment, the elimination module comprises:

And an operation submodule, configured to perform subtraction on the voice signal and the voice feedback signal to obtain the voice input signal.

In an embodiment, the transmission module comprises:

a conversion submodule, configured to perform analog-to-digital conversion on the voice input signal to obtain a digital voice input signal;

a processing submodule, configured to perform preset processing on the digital voice input signal, where the preset processing includes any one or more of a noise reduction process, a reverberation elimination process, and an echo cancellation process;

And a transmission submodule, configured to transmit the processed digital voice input signal to the voice recognition processor.

In an embodiment, the acquisition module comprises:

a receiving submodule, configured to receive an acquisition instruction sent by the voice recognition processor, where the collection instruction includes a sampling clock of a reset signal and a voice signal;

The collecting submodule is configured to collect a voice signal according to the collecting instruction.

In one embodiment, the apparatus further includes:

An identification module, configured to identify the voice signal before acquiring the voice signal;

a determining module, configured to determine, according to a preset effective voice signal feature, whether the voice signal is a valid voice signal;

An output module, configured to output an execution instruction to the voice recognition processor when the voice signal is the valid voice signal, where the execution instruction is used to instruct the voice recognition processor to issue an acquisition instruction, the collection instruction And used to instruct the sound collecting circuit to collect a voice signal.

Some beneficial effects of embodiments of the present invention may include:

The above device collects the voice input signal input by the user and the voice feedback signal broadcasted by the voice recognition processor, and eliminates the voice feedback signal therein, so that the finally collected voice signal is not interfered by the feedback signal, thereby obtaining “clean”. The voice input signal improves the accuracy of speech recognition.

In a third aspect, a pickup circuit is provided, including a microphone array, a digital signal processor, a voice broadcast interface, and a language Sound output interface; among them,

The microphone array is connected to the digital signal processor and configured to collect a voice signal, where the voice signal includes a voice input signal input by a user and a voice feedback reported by a voice recognition processor received through the voice broadcast interface. signal;

The digital signal processor is connected to the microphone array, the voice broadcast interface, and the voice output interface, and is configured to cancel a voice feedback signal in the voice signal to obtain the voice input signal;

The voice broadcast interface is connected to the digital signal processor and configured to receive a voice feedback signal broadcast by the voice recognition processor;

The voice output interface is coupled to the digital signal processor for transmitting the voice input signal to the voice recognition processor, and the voice recognition processor is configured to identify and report the voice input signal The speech feedback signal.

In an embodiment, the sound collecting circuit further includes:

a control unit, coupled to the digital signal processor, configured to receive an acquisition instruction sent by the speech recognition processor, the acquisition instruction includes a sampling clock of a reset signal and a voice signal; and control the microphone according to the acquisition instruction The array collects speech signals.

Some beneficial effects of embodiments of the present invention may include:

The above-mentioned sound collecting circuit simultaneously collects the voice input signal input by the user and the voice feedback signal broadcasted by the voice recognition processor through the microphone array, and eliminates the voice feedback signal therein by the digital signal processor, so that the finally collected voice signal is not subjected to feedback. Signal interference, resulting in a "clean" voice input signal, improving the accuracy of speech recognition.

According to a fourth aspect, a voice processing device is provided, which is applied to a sound pickup circuit, and the device includes:

processor;

a memory for storing the processor executable instructions;

Wherein the processor is configured to:

Acquiring a voice signal, the voice signal including a voice input signal input by the user and a voice feedback signal broadcast by the voice recognition processor;

Eliminating a voice feedback signal in the voice signal to obtain the voice input signal;

Transmitting the speech input signal to the speech recognition processor, the speech recognition processor for identifying the speech input signal and broadcasting the speech feedback signal.

The above processor is also configured to:

The speech signal and the speech feedback signal are subtracted to obtain the speech input signal.

The above processor is also configured to:

Performing analog-to-digital conversion on the voice input signal to obtain a digital voice input signal;

Performing a preset process on the digital voice input signal, the preset process including any one or more of a noise reduction process, a canceling reverberation process, and an echo cancellation process;

Transmitting the processed digital speech input signal to the speech recognition processor.

The above processor is also configured to:

Receiving an acquisition instruction sent by the voice recognition processor, where the acquisition instruction includes a sampling clock of a reset signal and a voice signal;

Acquiring a voice signal according to the acquisition instruction.

The above processor is also configured to:

Identifying the voice signal;

Determining whether the voice signal is a valid voice signal according to a preset effective voice signal feature;

And when the voice signal is the valid voice signal, outputting an execution instruction to the voice recognition processor, the execution instruction is used to instruct the voice recognition processor to issue an acquisition instruction, where the collection instruction is used to indicate the The sound pickup circuit collects a voice signal.

In a fifth aspect, there is provided a non-transitory computer readable recording medium having recorded thereon a computer program, the program comprising instructions for performing the method of the first aspect of the embodiment of the present invention.

In a sixth aspect, a computer program is provided, the program comprising: instructions for performing the method of the first aspect of the embodiments of the invention when the program is executed by a computer.

Other features and advantages of the invention will be set forth in the description which follows, The objectives and other advantages of the invention may be realized and obtained by means of the structure particularly pointed in the appended claims.

The technical solution of the present invention will be further described in detail below through the accompanying drawings and embodiments.

DRAWINGS

The drawings are intended to provide a further understanding of the invention, and are intended to be a In the drawing:

FIG. 1 is a flowchart of a voice processing method according to an embodiment of the present invention;

2 is a circuit block diagram of a sound pickup circuit and a voice recognition processor according to an embodiment of the present invention;

3 is a timing diagram of signals in an embodiment of the present invention;

4 is a flowchart of step S13 in a voice processing method according to an embodiment of the present invention;

4(a) is a timing diagram of a reset signal in an embodiment of the present invention;

FIG. 5 is a flowchart of a voice processing method according to an embodiment of the present invention;

FIG. 5(a) is a timing diagram of a wake-up signal according to an embodiment of the present invention; FIG.

FIG. 6 is a block diagram of a voice processing apparatus according to an embodiment of the present invention; FIG.

7 is a block diagram of a cancellation module in a voice processing device according to an embodiment of the present invention;

8 is a block diagram of a transmission module in a voice processing device according to an embodiment of the present invention;

FIG. 9 is a block diagram of an acquisition module in a voice processing device according to an embodiment of the present invention; FIG.

FIG. 10 is a block diagram of a voice processing apparatus according to an embodiment of the present invention; FIG.

11 is a block diagram of a sound pickup circuit according to an embodiment of the present invention;

FIG. 12 is a block diagram of a voice processing system according to an embodiment of the present invention; FIG.

FIG. 13 is a block diagram of an apparatus for performing a voice processing method according to an embodiment of the present invention.

detailed description

The preferred embodiments of the present invention are described with reference to the accompanying drawings, which are intended to illustrate and illustrate the invention.

FIG. 1 is a flowchart of a voice processing method according to an embodiment of the present invention. As shown in FIG. 1, the method is applied to a sound pickup circuit, and includes the following steps S11-S13:

In step S11, a voice signal is collected, where the voice signal includes a voice input signal input by the user and a voice feedback signal broadcast by the voice recognition processor.

In this step, the voice feedback signal broadcast by the voice recognition processor is a feedback signal to the voice input signal. Figure 2 shows a circuit block diagram of the pickup circuit and the speech recognition processor. As the voice signal is collected, the backend speech recognition processor 22 broadcasts the voice feedback signal through the speaker 23. The speaker 23 forms a loop with the microphone array 211 that collects the voice signal, as shown in the dotted circuit of FIG. The voice feedback signal broadcast by the recognition processor 22 through the speaker 23 causes interference to the voice signal. Therefore, the voice recognition processor 22 outputs the voice feedback signal to the sound pickup circuit 21 while the voice is being broadcast through the speaker 23. The circuit 21 eliminates the speech feedback signal causing the interference, thereby eliminating the interference of the speech feedback signal on the speech signal.

The sound collecting circuit 21 receives the voice feedback signal output by the voice recognition processor 23 through the voice broadcast interface 22, and the voice broadcast interface 22 can be an IIS (Inter-IC Sound) integrated bus mode. The signal name is defined as shown in Table 1, and the corresponding timing diagram is shown in Figure 3.

Table 1

Signal name	Signal direction	Signal description
SCLK	Output	Bit sampling clock
LRCK	Output	Left and right channel sync clock
SDI	Input	input signal
SDO	Output	output signal

Step S12, canceling the voice feedback signal in the voice signal to obtain a voice input signal.

In an embodiment, the step may be specifically implemented by: subtracting the voice signal and the voice feedback signal to obtain a voice input signal.

In step S13, the voice input signal is transmitted to the voice recognition processor, and the voice recognition processor is configured to identify the voice input signal and broadcast the voice feedback signal.

In this step, the sound collecting circuit transmits the voice input signal to the voice recognition processor through the voice output interface, wherein the voice output interface can be an IIS (Inter-IC Sound) bus mode, and the signal of the interface The name definition is shown in Table 1, and the corresponding timing diagram is shown in Figure 3.

According to the technical solution provided by the embodiment of the present invention, the voice input signal input by the user and the voice feedback signal broadcast by the voice recognition processor are simultaneously collected, and the voice feedback signal is eliminated, so that the finally collected voice signal is not affected by the feedback signal. Interference, resulting in a "clean" voice input signal, improving the accuracy of speech recognition.

In an embodiment, as shown in FIG. 4, step S13 may be implemented as the following steps S131-S133:

Step S131, performing analog-to-digital conversion on the voice input signal to obtain a digital voice input signal.

Step S132, performing preset processing on the digital voice input signal, and the preset processing includes any one or more of noise reduction processing, elimination of reverberation processing, and cancellation of echo processing.

Step S133, the processed digital voice input signal is transmitted to the voice recognition processor.

In this embodiment, the sound pickup circuit transmits the processed digital voice input signal to the voice recognition processor through the voice output interface. The sound pickup circuit performs analog-to-digital conversion on the voice input signal, as well as processing such as noise reduction, reverberation elimination, and echo cancellation, so that the voice recognition signal received by the voice recognition processor has no noise interference, thereby more accurately recognizing the voice input. signal.

In an embodiment, step S11 may be implemented as: receiving an acquisition instruction sent by the speech recognition processor, the acquisition instruction comprising a reset signal and a sampling clock of the speech signal; and acquiring the speech signal according to the acquisition instruction. In this embodiment, the voice signal can be collected according to the acquisition instruction sent by the voice recognition processor, and the reset control of the circuit is realized. That is, after the system is powered on, the speech recognition processor of the back end generates an acquisition instruction including a reset signal, and after receiving the acquisition instruction, the pickup circuit starts to collect the speech signal, that is, it is in a normal working state. The timing diagram of the reset signal RESET is shown in Figure 4(a).

In an embodiment, as shown in FIG. 5, before step S11, the above method further includes the following steps S51-S53:

In step S51, the voice signal is identified.

Step S52: Determine whether the voice signal is a valid voice signal according to a preset effective voice signal feature. If the voice signal is a valid voice signal, step S53 is performed; otherwise, returning to step S51, the newly received voice signal is continuously identified.

The effective voice signal feature can be set according to the voice service performed by the current system. For example, if the current system performs banking, the effective voice signal feature may be set to include keywords involved in the banking service. When the voice signal is received, the sound collecting circuit identifies the keyword according to the received voice signal. When the keyword is a keyword involved in banking, the voice signal can be determined to be a valid voice signal.

Step S53, outputting an execution instruction to the voice recognition processor, the execution instruction is used to instruct the voice recognition processor to issue an acquisition instruction, and the collection instruction is used to instruct the sound pickup circuit to collect the voice signal.

Since the voice recognition processor is in a sleep state when no valid voice signal is received, in the embodiment, the issuance of the execution instruction actually implements the wake-up function of the sound pickup circuit to the voice recognition processor, that is, equivalent A wake-up signal is issued, and the timing chart of the wake-up signal WAKE_INT is as shown in FIG. 5(a).

In this embodiment, by recognizing the received voice signal, and instructing the voice recognition processor to issue an acquisition instruction when the voice signal is a valid voice signal, the voice recognition processor wakes up the sound pickup circuit to avoid picking up The tone circuit receives an invalid voice signal.

Corresponding to the voice processing method in the above embodiment, the present invention further provides a voice processing device for performing the above method.

FIG. 6 is a block diagram of a voice processing apparatus according to an embodiment of the present invention. As shown in FIG. 6, the device is applied to a sound pickup circuit, and includes:

The acquiring module 61 is configured to collect a voice signal, where the voice signal includes a voice input signal input by the user and a voice feedback signal broadcast by the voice recognition processor;

The eliminating module 62 is configured to cancel the voice feedback signal in the voice signal to obtain a voice input signal;

The transmission module 63 is configured to transmit the voice input signal to the voice recognition processor, and the voice recognition processor is configured to identify the voice input signal and broadcast the voice feedback signal.

In one embodiment, as shown in FIG. 7, the elimination module 62 includes:

The operation sub-module 621 is configured to perform a subtraction operation on the voice signal and the voice feedback signal to obtain a voice input signal.

In one embodiment, as shown in FIG. 8, the transmission module 63 includes:

a conversion sub-module 631, configured to perform analog-to-digital conversion on the voice input signal to obtain a digital voice input signal;

The processing sub-module 632 is configured to perform preset processing on the digital voice input signal, where the preset processing includes any one or more of noise reduction processing, anti-reverberation processing, and echo cancellation processing;

The transmission sub-module 633 is configured to transmit the processed digital voice input signal to the voice recognition processor.

In an embodiment, as shown in FIG. 9, the acquisition module 61 includes:

The receiving sub-module 611 is configured to receive an acquisition instruction sent by the voice recognition processor, where the collection instruction includes a sampling clock of the reset signal and the voice signal;

The collecting sub-module 612 is configured to collect a voice signal according to the acquisition instruction.

In an embodiment, as shown in FIG. 10, the foregoing apparatus further includes:

The identification module 64 is configured to identify the voice signal before acquiring the voice signal;

The determining module 65 is configured to determine, according to the preset effective voice signal feature, whether the voice signal is a valid voice signal;

The output module 66 is configured to output an execution instruction to the voice recognition processor when the voice signal is a valid voice signal, the execution instruction is used to instruct the voice recognition processor to issue an acquisition instruction, and the collection instruction is used to instruct the sound collection circuit to collect the voice signal.

By using the device provided by the embodiment of the present invention, the voice input signal input by the user and the voice feedback signal broadcast by the voice recognition processor are simultaneously collected, and the voice feedback signal is eliminated, so that the finally collected voice signal is not interfered by the feedback signal. In order to get a "clean" voice input signal, the accuracy of speech recognition is improved.

Figure 11 is a block diagram of a sound pickup circuit in accordance with an embodiment of the present invention. As shown in FIG. 11, the sound collecting circuit includes a microphone array 111, a digital signal processor 112, a voice broadcast interface 113, and a voice output interface 114;

The microphone array 111 is connected to the digital signal processor 112 for collecting a voice signal. The voice signal includes a voice input signal input by the user and a voice feedback signal broadcast by the voice recognition processor received through the voice broadcast interface 113.

The microphone array 111 uses an array of multiple microphones to eliminate interference noise in different directions. It can usually be composed of 2 mics, 4 mics or 6 mics, and the mic array 111 is uniformly distributed according to the corresponding angles.

The digital signal processor 112 is connected to the microphone array 111, the voice broadcast interface 113, and the voice output interface 114. It is used to cancel the speech feedback signal in the speech signal and obtain the speech input signal.

The digital signal processor 112 includes an analog to digital conversion component for analog to digital conversion of the received speech input signal to obtain a digital speech input signal. The digital signal processor 112 performs a preset process on the digital voice input signal, the preset process including any one or more of a noise reduction process, a reverberation canceling process, and an echo cancellation process, and the processing is performed through the voice output interface 114. The digital voice input signal is transmitted to the speech recognition processor.

The voice broadcast interface 113 is connected to the digital signal processor 112 for receiving the voice feedback signal broadcast by the voice recognition processor.

The voice broadcast interface 113 may be in the form of an IIS (Inter-IC Sound) bus.

The voice output interface 114 is coupled to the digital signal processor 112 for transmitting the voice input signal to the voice recognition processor, the voice recognition processor for identifying the voice input signal and broadcasting the voice feedback signal.

The voice output interface 114 can be in the form of an IIS (Inter-IC Sound) bus.

The working principle of the sound collecting circuit is as follows: First, the microphone array 111 collects the voice input signal input by the user, and simultaneously collects the voice feedback signal output by the voice recognition processor through the voice broadcast interface 113; secondly, the digital signal processor 112 pairs the microphone array 111. The collected voice signal is processed, the voice feedback signal is eliminated, a "clean" voice input signal is obtained, and the "clean" voice input signal is transmitted through the voice output interface 114 to the voice recognition processor at the back end to make the voice The recognition processor is capable of recognizing "clean" speech input signals, thereby improving the accuracy of speech recognition.

In an embodiment, the sound collecting circuit further includes:

The control unit is connected to the digital signal processor 112 for receiving an acquisition instruction sent by the voice recognition processor, and the acquisition instruction includes a sampling clock of the reset signal and the voice signal; and controlling the microphone array 111 to collect the voice signal according to the acquisition instruction. In addition, the control component can also be used for transmission of control commands between the digital signal processor 112 and the speech recognition processor, for example, to increase the front end gain of the microphone array, to adjust the speech output format, and the like.

The interface of the control component may be an IIC (Inter-Integrated Circuit) bus interface, an SPI (Serial Peripheral Interface) or a UART (Universal Asynchronous Receiver/Transmitter) bus interface.

In one embodiment, the digital signal processor 112 is further configured to identify the voice signal; determine whether the voice signal is a valid voice signal according to a preset effective voice signal feature; and pass the control component when the voice signal is a valid voice signal And outputting an execution instruction to the voice recognition processor, the execution instruction is used to instruct the voice recognition processor to issue an acquisition instruction, and the collection instruction is used to instruct the sound collection circuit to acquire the voice signal.

FIG. 12 is a block diagram of a voice processing system according to an embodiment of the present invention. As shown in Figure 12, the voice processing system package A sound pickup circuit 121 and a voice recognition processor 122 are included. The sound collecting circuit 121 is configured to collect a voice signal, process the voice signal, and then transmit the processed voice signal to the voice recognition processor 122, and the voice recognition processor 122 performs a recognition process on the voice signal.

In an embodiment, after the voice processing system is powered on, the voice recognition processor 122 generates an acquisition instruction, and sends the collection instruction to the sound collection circuit 121, where the collection instruction includes a sampling clock of the reset signal and the voice signal. . The reset signal is used to control the sound collecting circuit 121 to enter an operating state, and start collecting voice signals. The speech recognition processor 122 sends an acquisition instruction to the pickup circuit 121 through a corresponding hardware interface to implement reset control.

In one embodiment, when a valid voice signal is not received, the voice recognition processor 122 is in a sleep state, and the collected voice signal is matched by the digital signal processor in the sound pickup circuit 121 to determine the collected voice. Whether the signal is a valid speech signal, when it is a valid signal, the pickup circuit 121 issues an execution instruction to the speech recognition processor 122 to wake up the speech recognition processor 122. The pickup circuit 121 transmits an execution instruction to the voice recognition processor 122 through a corresponding hardware interface to implement a wake-up function to the voice recognition processor 122.

FIG. 13 is a block diagram of an apparatus for performing a voice processing method, according to an exemplary embodiment. For example, device 1600 can be a mobile phone, a computer, a digital broadcast terminal, a messaging device, a gaming console, a tablet device, a medical device, a fitness device, a personal digital assistant, and the like.

Referring to Figure 13, device 1600 can include one or more of the following components: processor 1601, memory 1602, and communication component 1603.

The processor 1601 typically controls the overall operation of the device 1600, such as operations associated with display, telephone calls, data communications, camera operations, and recording operations. The processor 1601 can execute instructions to perform all or part of the steps of the above method.

Memory 1602 is configured to store various types of data to support operation at device 1600. Examples of such data include instructions for any application or method operating on device 1600, contact data, phone book data, messages, pictures, videos, and the like. The memory 1602 can be implemented by any type of volatile or non-volatile storage device, or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read only memory (EEPROM), erasable. Programmable Read Only Memory (EPROM), Programmable Read Only Memory (PROM), Read Only Memory (ROM), Magnetic Memory, Flash Memory, Disk or Optical Disk.

Communication component 1603 is configured to facilitate wired or wireless communication between device 1600 and other devices. The device 1600 can access a wireless network based on a communication standard, such as Wi-Fi, 2G or 3G, or a combination thereof. In an exemplary embodiment, the communication component 1603 receives a broadcast signal or broadcast associated information from an external broadcast management system via a broadcast channel. In an exemplary embodiment, communication component 1603 also includes a near field communication (NFC) module to facilitate short Cheng Communication. For example, the NFC module can be implemented based on radio frequency identification (RFID) technology, infrared data association (IrDA) technology, ultra-wideband (UWB) technology, Bluetooth (BT) technology, and other technologies.

In an exemplary embodiment, device 1600 may be implemented by one or more application specific integrated circuits (ASICs), digital signal processors (DSPs), digital signal processing devices (DSPDs), programmable logic devices (PLDs), field programmable A gate array (FPGA), controller, microcontroller, microprocessor or other electronic component implementation for performing the voice processing method described above.

In an exemplary embodiment, there is also provided a non-transitory computer readable storage medium comprising instructions, such as a memory 1602 comprising instructions executable by processor 1601 of apparatus 1600 to perform the voice processing method described above. For example, the non-transitory computer readable storage medium can be a ROM, a random access memory (RAM), a CD-ROM, a magnetic tape, a floppy disk, and an optical data storage device.

The present invention also provides a non-transitory computer readable recording medium having recorded thereon a computer program including instructions for executing the voice processing method according to the above-described embodiments of the present invention.

The present invention also provides a computer program comprising: instructions for executing a speech processing method according to the above-described embodiments of the present invention when the program is executed by a computer.

Those skilled in the art will appreciate that embodiments of the present invention can be provided as a method, system, or computer program product. Accordingly, the present invention may take the form of an entirely hardware embodiment, an entirely software embodiment, or a combination of software and hardware. Moreover, the invention can take the form of a computer program product embodied on one or more computer-usable storage media (including but not limited to disk storage and optical storage, etc.) including computer usable program code.

The present invention has been described with reference to flowchart illustrations and/or block diagrams of methods, apparatus (system), and computer program products according to embodiments of the invention. It will be understood that each flow and/or block of the flowchart illustrations and/or FIG. These computer program instructions can be provided to a processor of a general purpose computer, special purpose computer, embedded processor, or other programmable data processing device to produce a machine for the execution of instructions for execution by a processor of a computer or other programmable data processing device. Means for implementing the functions specified in one or more of the flow or in a block or blocks of the flow chart.

The computer program instructions can also be stored in a computer readable memory that can direct a computer or other programmable data processing device to operate in a particular manner, such that the instructions stored in the computer readable memory produce an article of manufacture comprising the instruction device. The apparatus implements the functions specified in one or more blocks of a flow or a flow and/or block diagram of the flowchart.

These computer program instructions can also be loaded onto a computer or other programmable data processing device such that the computer Or performing a series of operational steps on other programmable devices to produce computer-implemented processing such that instructions executed on a computer or other programmable device are provided for implementing a block in a flow or a flow and/or block diagram of the flowchart Or the steps of the function specified in multiple boxes.

It is apparent that those skilled in the art can make various modifications and variations to the invention without departing from the spirit and scope of the invention. Thus, it is intended that the present invention cover the modifications and modifications of the invention

Claims (18)

1. A voice processing method is characterized in that it is applied to a sound pickup circuit, and the method includes:

   Acquiring a voice signal, the voice signal including a voice input signal input by the user and a voice feedback signal broadcast by the voice recognition processor;

   Eliminating a voice feedback signal in the voice signal to obtain the voice input signal;

   Transmitting the speech input signal to the speech recognition processor, the speech recognition processor for identifying the speech input signal and broadcasting the speech feedback signal.
2. The method according to claim 1, wherein the canceling the voice feedback signal in the voice signal to obtain the voice input signal comprises:

   The speech signal and the speech feedback signal are subtracted to obtain the speech input signal.
3. The method of claim 1, wherein the transmitting the voice input signal to the voice recognition processor comprises:

   Performing analog-to-digital conversion on the voice input signal to obtain a digital voice input signal;

   Performing a preset process on the digital voice input signal, the preset process including any one or more of a noise reduction process, a canceling reverberation process, and an echo cancellation process;

   Transmitting the processed digital speech input signal to the speech recognition processor.
4. The method according to claim 1, wherein the acquiring the voice signal comprises:

   Receiving an acquisition instruction sent by the voice recognition processor, where the acquisition instruction includes a sampling clock of a reset signal and a voice signal;

   Acquiring a voice signal according to the acquisition instruction.
5. The method according to claim 1, wherein before the acquiring the voice signal, the method further comprises:

   Identifying the voice signal;

   Determining whether the voice signal is a valid voice signal according to a preset effective voice signal feature;

   And when the voice signal is the valid voice signal, outputting an execution instruction to the voice recognition processor, the execution instruction is used to instruct the voice recognition processor to issue an acquisition instruction, where the collection instruction is used to indicate the The sound pickup circuit collects a voice signal.
6. The method according to claim 4, wherein before the acquiring the voice signal, the method further comprises:

   Identifying the voice signal;

   Determining whether the voice signal is a valid voice signal according to a preset effective voice signal feature;

   And when the voice signal is the valid voice signal, outputting an execution instruction to the voice recognition processor, the execution instruction is used to instruct the voice recognition processor to issue an acquisition instruction, where the collection instruction is used to indicate the The sound pickup circuit collects a voice signal.
7. A sound pickup circuit, comprising: a microphone array, a digital signal processor, a voice broadcast interface, and a voice output interface; wherein

   The microphone array is connected to the digital signal processor and configured to collect a voice signal, where the voice signal includes a voice input signal input by a user and a voice feedback reported by a voice recognition processor received through the voice broadcast interface. signal;

   The digital signal processor is connected to the microphone array, the voice broadcast interface, and the voice output interface, and is configured to cancel a voice feedback signal in the voice signal to obtain the voice input signal;

   The voice broadcast interface is connected to the digital signal processor and configured to receive a voice feedback signal broadcast by the voice recognition processor;

   The voice output interface is coupled to the digital signal processor for transmitting the voice input signal to the voice recognition processor, and the voice recognition processor is configured to identify and report the voice input signal The speech feedback signal.
8. The sound pickup circuit according to claim 7, wherein said digital signal processor is configured to subtract said speech signal and said speech feedback signal to obtain said speech input signal.
9. The sound pickup circuit according to claim 8, wherein said voice output interface is configured to:

   Performing analog-to-digital conversion on the voice input signal to obtain a digital voice input signal;

   Performing a preset process on the digital voice input signal, the preset process including any one or more of a noise reduction process, a canceling reverberation process, and an echo cancellation process;

   Transmitting the processed digital speech input signal to the speech recognition processor.
10. The sound pickup circuit according to claim 8, wherein said microphone array is configured to:

    Receiving an acquisition instruction sent by the voice recognition processor, where the acquisition instruction includes a sampling clock of a reset signal and a voice signal;

    Acquiring a voice signal according to the acquisition instruction.
11. The sound collecting circuit according to claim 8, wherein the sound collecting circuit further identifies the voice signal before acquiring the voice signal; and determining whether the voice signal is based on a preset effective voice signal characteristic Is an effective voice signal; when the voice signal is the valid voice signal, outputting to the voice recognition processor a line instruction, the execution instruction is used to instruct the voice recognition processor to issue an acquisition instruction, and the collection instruction is used to instruct the sound collection circuit to acquire a voice signal.
12. The sound collecting circuit according to claim 10, wherein the sound collecting circuit further identifies the voice signal before acquiring the voice signal; and determining whether the voice signal is based on a preset effective voice signal characteristic An effective voice signal; when the voice signal is the valid voice signal, outputting an execution instruction to the voice recognition processor, the execution instruction being used to instruct the voice recognition processor to issue an acquisition instruction, the collection instruction And used to instruct the sound collecting circuit to collect a voice signal.
13. A non-transitory computer readable recording medium having recorded thereon a computer program, the program comprising instructions for executing a voice processing method, the method comprising:

    Acquiring a voice signal, the voice signal including a voice input signal input by the user and a voice feedback signal broadcast by the voice recognition processor;

    Eliminating a voice feedback signal in the voice signal to obtain the voice input signal;

    Transmitting the speech input signal to the speech recognition processor, the speech recognition processor for identifying the speech input signal and broadcasting the speech feedback signal.
14. The computer readable recording medium according to claim 13, wherein the canceling the voice feedback signal in the voice signal to obtain the voice input signal comprises:

    The speech signal and the speech feedback signal are subtracted to obtain the speech input signal.
15. The computer readable recording medium according to claim 13, wherein said transmitting said voice input signal to said voice recognition processor comprises:

    Performing analog-to-digital conversion on the voice input signal to obtain a digital voice input signal;

    Performing a preset process on the digital voice input signal, the preset process including any one or more of a noise reduction process, a canceling reverberation process, and an echo cancellation process;

    Transmitting the processed digital speech input signal to the speech recognition processor.
16. The computer readable recording medium according to claim 13, wherein the acquiring the voice signal comprises:

    Receiving an acquisition instruction sent by the voice recognition processor, where the acquisition instruction includes a sampling clock of a reset signal and a voice signal;

    Acquiring a voice signal according to the acquisition instruction.
17. The computer readable recording medium according to claim 13, wherein before the acquiring the voice signal, the method further comprises:

    Identifying the voice signal;

    Determining whether the voice signal is a valid voice signal according to a preset effective voice signal feature;

    And when the voice signal is the valid voice signal, outputting an execution instruction to the voice recognition processor, the execution instruction is used to instruct the voice recognition processor to issue an acquisition instruction, where the collection instruction is used to indicate the The sound pickup circuit collects a voice signal.
18. The computer readable recording medium according to claim 16, wherein before the acquiring the voice signal, the method further comprises:

    Identifying the voice signal;

    Determining whether the voice signal is a valid voice signal according to a preset effective voice signal feature;

    And when the voice signal is the valid voice signal, outputting an execution instruction to the voice recognition processor, the execution instruction is used to instruct the voice recognition processor to issue an acquisition instruction, where the collection instruction is used to indicate the The sound pickup circuit collects a voice signal.

Images