阅读说明：本技术 信号处理装置、信号处理方法和信号处理程序 (Signal processing device, signal processing method, and signal processing program ) 是由 林繁利 浅田宏平 土谷慎平 大栗一敦 于 2019-03-08 设计创作，主要内容包括：一种信号处理装置,通过连接多个单元来执行噪声消除处理,该信号处理装置配备有噪声消除处理单元,一个或多个输入单元和一个或多个输出单元可以连接到该噪声消除处理单元。(A signal processing apparatus which performs noise cancellation processing by connecting a plurality of units is provided with a noise cancellation processing unit to which one or more input units and one or more output units can be connected.)

1. A signal processing apparatus comprising:

a noise cancellation processing unit connectable to the one or more input units and connectable to the one or more output units, the plurality of signal processing devices being connected to each other and configured to perform noise cancellation processing.

2. The signal processing apparatus of claim 1, wherein

The plurality of signal processing devices are daisy-chained.

3. The signal processing apparatus of claim 1, wherein

Data is transmitted between a plurality of the noise canceling processing units.

4. A signal processing apparatus according to claim 3, wherein

The data is an audio signal input from the one or more input units.

5. A signal processing apparatus according to claim 3, wherein

The data is a cancellation signal output from the one or more output units.

6. A signal processing apparatus according to claim 3, wherein

The data is control information.

7. The information processing apparatus according to claim 3, wherein

Reducing the size of the data and transmitting the data.

8. The information processing apparatus according to claim 7, wherein

The size of the data is reduced by reducing the sampling frequency.

9. The information processing apparatus according to claim 7, wherein

The size of the data is reduced by reducing the bit rate.

10. The signal processing apparatus of claim 1, wherein

One of the plurality of noise canceling processing units to which an input unit close to a noise source is connected transmits data to another one of the plurality of noise canceling processing units to which an input unit far from the noise source is connected.

11. The signal processing apparatus of claim 1, wherein

The signal processing device is connected to a noise analyzer unit that analyzes noise, and switches the noise cancellation process according to an analysis result of the noise analyzer unit.

12. The signal processing apparatus of claim 11, wherein

The signal processing device switches the mode of the noise cancellation processing according to the analysis result of the noise analyzer unit.

13. The signal processing apparatus of claim 11, wherein

The noise cancellation processing unit may perform the noise cancellation processing in a plurality of ways, and change a combination of the ways according to an analysis result of the noise analyzer unit.

14. The signal processing apparatus of claim 1, wherein

The plurality of input units and the plurality of output units are connected to any one of the plurality of noise cancellation processing units connected to each other.

15. A signal processing method, comprising:

a plurality of signal processing apparatuses are connected to each other and perform noise cancellation processing, each of the plurality of signal processing apparatuses including a noise cancellation processing unit connectable to one or more input units and connectable to one or more output units.

16. A signal processing program that causes a computer to execute a signal processing method comprising:

a plurality of signal processing apparatuses are connected to each other and perform noise cancellation processing, each of the plurality of signal processing apparatuses including a noise cancellation processing unit connectable to one or more input units and connectable to one or more output units.

Technical Field

The present technology relates to a signal processing apparatus, a signal processing method, and a signal processing program.

Background

Conventionally, a noise cancellation technique for reducing noise in a space by using a predetermined number of speakers and microphones has been proposed (patent document 1).

Further, in noise control in a specific closed space, it is known to improve noise reduction performance by using a system configuration that takes into account mutual interference between multiple inputs and multiple outputs (multiple inputs-multiple outputs). This is different from the single input and single output seen in headphone noise cancellation.

CITATION LIST

Patent document

Patent document 1: japanese patent application laid-open No. 2015-080199.

Disclosure of Invention

Technical problem

However, in consideration of the size of a space to be controlled and the resource of signal processing, it is inefficient to implement a configuration of multiple inputs and multiple outputs in a single noise canceling system. Meanwhile, the configuration of multiple inputs and multiple outputs has a problem that the system scale becomes large.

The present technology has been proposed to address such a problem. An object of the present technology is to provide a signal processing device capable of easily adjusting the scale of a target range of noise cancellation processing. An object of the present technology is to provide a signal processing method and a signal processing program.

Solution to the problem

In order to solve the above-described problem, according to a first technique, there is provided a noise cancellation processing unit connectable to one or more input units and connectable to one or more output units, a plurality of signal processing devices being connected to each other and configured to perform noise cancellation processing.

Further, according to a second technique, there is provided a signal processing method including: a plurality of signal processing devices are connected to each other and perform noise cancellation processing, each of the plurality of signal processing devices including a noise cancellation processing unit connectable to one or more input units and connectable to one or more output units.

Further, according to a third technique, there is provided a signal processing program that causes a computer to execute a signal processing method including connecting a plurality of signal processing apparatuses to each other and performing noise cancellation processing, each of the plurality of signal processing apparatuses including a noise cancellation processing unit connectable to one or more input units and connectable to one or more output units.

The invention has the advantages of

According to the present technology, the scale of the target range of the noise cancellation process can be easily adjusted. It should be noted that the effects of the present technology are not limited to those described herein. The present techniques may have any of the effects described herein.

Drawings

Fig. 1 shows a block diagram of a configuration of a signal processing apparatus according to an embodiment of the present technology.

Fig. 2 shows a diagram of a first feedback system.

Fig. 3 shows a diagram of a second feedback system.

Fig. 4 shows a diagram of a third feedback system.

Fig. 5 shows a diagram of the connection of the signal processing means of the feedforward system.

Fig. 6 shows a diagram of the connection of the signal processing means of the feedback system.

FIG. 7 is a diagram for explaining the connection of a signal processing device of a feedforward system and a signal processing device of a feedback system.

Fig. 8 is a diagram for explaining the connection of the signal processing device of the first feedback system and the signal processing device of the second feedback system.

Fig. 9 is a diagram for explaining the connection of the signal processing device of the feedforward system and the signal processing device of the third feedback system.

Fig. 10 is a diagram for explaining the arrival direction of noise from a noise source.

Fig. 11 is an explanatory diagram of a connection situation of a signal processing apparatus as a first feedback system of the 8-shaped loop canceller.

Fig. 12 shows a diagram of the connection of the noise analyzer to the signal processing device.

Fig. 13 shows a diagram of a case where modules are arranged in a circular array and a noise source exists outside the circular array.

Fig. 14 is an explanatory diagram of data transmission in the example of fig. 13.

Fig. 15 shows a diagram in which modules are arranged in a circular array, where there is a noise source in the circular array.

Fig. 16 shows a diagram of data transmission in the example of fig. 15.

Fig. 17 shows a table of the format of data to be transmitted.

Fig. 18 is a diagram for explaining the direction of data transmission.

Fig. 19 is a diagram showing a first example of a packet in data transmission.

Fig. 20 is a diagram showing a second example of a packet in data transmission.

Fig. 21 is a diagram showing data transmission in the module configuration shown in fig. 18.

Fig. 22 shows a diagram of an example of multiple-input and multiple-output processing using reference signals collected by reference microphones of two adjacent modules.

Fig. 23 shows a diagram of an example of multiple-input and multiple-output processing using reference signals collected by reference microphones of two adjacent modules.

Fig. 24 is a signal processing block diagram of a second feedback system in a multiple-input and multiple-output system.

Detailed Description

Embodiments of the present technology will be described below with reference to the drawings. Note that the description is given in the following order.

<1. example >

[1-1. arrangement of Signal processing Unit ]

[1-2. connection of Signal processing device ]

[1-3. data Transmission ]

[1-3-1. first example of circular array ]

[1-3-2. second example of circular array ]

[1-3-3. direction of data Transmission ]

[1-3-4. packet in data Transmission ]

<2. modification >

<1. example >

[1-1. arrangement of Signal processing Unit ]

The configuration of the signal processing apparatus 100 will be described first with reference to fig. 1. The signal processing apparatus 100 includes a noise cancellation processing unit 101 and a control unit 102. The plurality of microphones 111 are connected to the signal processing apparatus 100 via a plurality of AD (analog/digital) converters 113 and a plurality of microphone amplifiers 112. Further, a plurality of speakers 116 are connected via a plurality of DA (digital/analog) converters 114 and a plurality of power amplifiers 115.

Further, the sound source 130 is connected to the signal processing apparatus 100 via the digital I/F121. Note that the sound source 130 and the digital I/F121 are not necessarily connected to each other. Further, a synchronization circuit 140 is connected to the signal processing apparatus 100.

The plurality of microphones 111 are connected to the noise cancellation processing unit 101 via a plurality of microphone amplifiers 112 and a plurality of AD converters 113. Further, a plurality of speakers 116 are connected to the noise cancellation processing unit 101 via a plurality of DA converters 114 and a power amplifier 115. Thus, one or more inputs and one or more outputs may be connected to the noise cancellation processing unit 101. Thus, the signal processing apparatus 100 is configured as a multiple-input multiple-output apparatus. The signal processing apparatus 100 can reduce noise in a space (hereinafter referred to as a processing range) to be subjected to noise cancellation processing by using a plurality of inputs and a plurality of outputs.

The microphone 111 collects sound and noise within a processing range where noise reduction is performed by the signal processing apparatus 100. An audio signal based on the sound collection result of the microphone 111 is supplied to the AD converter 113, and the gain is adjusted by the microphone amplifier 112. The AD converter 113 converts an audio signal, which is an analog signal, into a digital signal, and supplies the digital signal to the noise cancellation processing unit 101. The microphone 111 corresponds to an input unit in claims.

The noise removal processing unit 101 includes a digital filter for generating a noise reduction audio signal (hereinafter, referred to as a removal signal). The noise removal processing unit 101 uses the supplied digital audio signal to generate a removal signal having a characteristic corresponding to the filter coefficient as a predetermined parameter. The noise removal processing unit 101 supplies the removal signal to the plurality of DA converters 114. Alternatively, the noise removal processing unit 101 may supply to the plurality of DA converters 114 by generating a removal signal obtained by inverting the phase of the supplied digital audio signal. The control unit 102 controls the entire signal processing apparatus 100 and each unit, and further controls and manages communication between the connected signal processing apparatuses 100. The noise removal processing unit 101 and the control unit 102 are each configured by a DSP (digital signal processing device) or the like.

The signal processing device 100 is configured by a program. The program may be installed in advance in a processor such as a DSP or a computer for performing signal processing. The program may be distributed via download, storage media, or the like to be installed by the user. Further, the signal processing apparatus 100 may be realized not only by a program but also by a combination of a dedicated device, a circuit, and the like of hardware having these functions.

The DA converter 114 converts the supplied cancellation signal into an analog signal. The DA converter 114 supplies the cancellation signal to the power amplifier 115. The power amplifier 115 then provides the cancellation signal to the speaker 116. The speaker 116 outputs the cancellation signal. Therefore, noise in the processing range can be reduced. The speaker 116 corresponds to an output unit in claims.

The sound source 130 may also provide the audio content signal to the noise cancellation processing unit 101 via the digital I/F121. The sound source 130 is a music player, a DVD player, a blu-ray (registered trademark) player, various media players (e.g., car stereo). The audio content signal provided from the sound source 130 is an audio signal reproduced by a media player. The user listens to the audio content signal as audio content within a processing range in which noise is removed by the signal processing apparatus 100.

When the user listens to the audio content from the sound source 130 within the processing range of the signal processing apparatus 100, the audio content reproduced from the sound source 130 and noise are input to the microphone 111 within the processing range. The noise removal processing unit 101 removes audio content from the audio content and the noise signal using the audio content signal supplied through the digital I/F121. Therefore, the noise removal processing unit 101 generates a signal of only noise. The noise removal processing unit 101 generates a removal signal from the signal of only noise, and outputs the removal signal from the speaker 116. Therefore, it is possible to reduce only noise without affecting audio content reproduced from the sound source 130 within a processing range.

The signal processing system includes a plurality of signal processing apparatuses 100 connected to each other. In this case, the synchronization circuit 140 generates and supplies a click signal for synchronizing all of the plurality of signal processing apparatuses 100 connected.

The plurality of signal processing apparatuses 100 thus configured are daisy-chained by a dedicated bus 150. Therefore, a signal processing system including a plurality of signal processing apparatuses 100 can be configured. Therefore, the scale of the signal processing system can be increased according to the size of the processing range that is the target of the noise cancellation processing. The communication over the dedicated bus 150 enables the transmission of various data such as control information, audio signals, cancellation signals, and the like.

The present technique is useful in any environment in order to reduce noise in space. For example, the present technology is applied to a room of a house. Therefore, noise entering the room from outside the house and noise generated in the room can be reduced. Then, the signal processing devices are daisy-chained according to the size of the room to adjust the scale of the signal processing system. Therefore, even in a large room, noise can be reduced appropriately. The present technology can also be applied to a vehicle to reduce noise from outside the vehicle. It is also possible to reduce noise generated inside the vehicle.

When the signal processing apparatus 100 is used in such a room or vehicle, there is a case where a speaker for audio content output and a speaker for canceling signal output are shared. In this case, only noise is reduced, and audio content output from the speaker is not reduced. For this purpose, the sound source 130 is connected to the signal processing device 100 via a digital I/F121. The sound source 130 supplies the audio content signal to the noise cancellation processing unit 101. Then, the noise removal processing unit 101 removes the audio content signal from the signal and noise of the audio content collected by the microphone. Therefore, the noise removal processing unit 101 generates a signal that is only noise. The noise cancellation processing unit 101 is used to generate a cancellation signal from a signal that is only noise. This enables only noise reduction without reducing the audio content from the sound source 130 within the processing range.

When a plurality of signal processing apparatuses 100 are connected through the dedicated bus 150 to form a signal processing system, audio content signals must also be transmitted between the signal processing apparatuses via the dedicated bus 150. As the audio content, a voice call of a telephone and a voice command may also be used.

In the following description, a module refers to a configuration in which a microphone amplifier, an AD converter, a DA converter, and a power amplifier are connected to a signal processing apparatus. A microphone and speaker are connected to the module.

Next, classification of the noise canceling system will be described. Noise cancellation systems can be largely divided into feedforward systems and feedback systems.

According to the feedforward system, noise is collected by a microphone to obtain a noise signal, predetermined signal processing is performed on the noise signal to generate a cancellation signal, and the cancellation signal is output from a speaker or the like. This reduces noise. According to this feed forward system, a reference microphone for collecting noise is required.

According to the feedback system, noise and sound reproduced in a processing range are collected by a microphone, only a noise component is extracted from an audio signal, and predetermined signal processing is performed on the audio signal to generate a cancel signal. Then, the cancellation signal is output from a speaker or the like. This reduces noise. According to this feedback system, an error microphone for acquiring and feeding back a noise reduction error (residual noise) is required.

In addition, there are a first feedback system, a second feedback system, and a third feedback system in the feedback system.

The first feedback system maximizes the denominator of the sensitivity function based on classical control engineering, as shown in fig. 2. This is a technique for reducing noise.

The second feedback system is a method of introducing an internal model into a feedback loop as shown in fig. 3 and minimizing the numerator of a sensitivity function to reduce noise.

The third feedback system is a combination of the first method and the second method, as shown in fig. 4.

These methods may be combined to enhance the performance of noise cancellation if more accurate noise cancellation processing is required.

[1-2. connection of Signal processing device ]

In fig. 5, a plurality of signal processing apparatuses 100 that perform noise cancellation of the feedforward system are daisy-chain connected. In this example, this constitutes a multiple-input multiple-output signal processing system. The signal processing apparatus 100 is connected by a dedicated bus 150. Therefore, the modules 210 and 220 … … are connected to form a multiple-input multiple-output signal processing system.

Further, in fig. 6, a plurality of signal processing apparatuses 100 are daisy-chained to perform feedback system noise cancellation. This is an example of constructing a multiple-input multiple-output signal processing system. The signal processing apparatus 100 is connected by a dedicated bus 150. Therefore, the modules 230 and 240 … … are connected to form an input and output signal processing system.

In this way, even in the case of the feedforward type noise cancellation or the feedback type noise cancellation, the plurality of signal processing apparatuses 100 are daisy-chained using the dedicated bus 150. As a result, the number of inputs and the number of outputs can be increased. Further, the number of noise canceling processing units 101 for performing noise canceling processing can be increased. As a result, the processing range in which noise cancellation can be performed can be expanded. Therefore, the scale of the signal processing system can be expanded according to the expansion of the processing range. In addition, noise cancellation performance can be improved.

In fig. 7, the feedforward signal processing means 100 and the first feedback signal processing means 100 are daisy-chained. In this example, a plurality of modules 310, 320 … … are connected to form a multiple-input multiple-output signal processing system. The microphone 111 connected to the feedforward signal processing apparatus 100 serves as a reference microphone for collecting noise. The microphone 111 of the signal processing device 100 connected to the first feedback system also serves as an error microphone for obtaining a noise reduction error.

In fig. 7, the signal processing apparatus 100 of the feedforward system and the signal processing apparatus 100 of the first feedback system are daisy-chain connected, and can transmit and receive a cancellation signal. Therefore, a speaker for outputting a cancellation signal may be connected to any one of the signal processing apparatuses 100. In fig. 7, a speaker 116 is connected to the signal processing apparatus 100 of the feedforward system. Alternatively, a speaker may be connected to the signal processing apparatus 100 of the first feedback system.

In fig. 8, the signal processing device 100 of the first feedback system and the signal processing device 100 of the second feedback system are daisy-chained. In this example, a plurality of modules 410, 420 … … are connected to form a multiple-input multiple-output signal processing system. The combination of the first feedback system and the second feedback system is an example of the third feedback system. The first feedback system and the second feedback system are both feedback systems. Therefore, the microphone and the speaker can be shared by the signal processing apparatus 100 of the first feedback system and the signal processing apparatus 100 of the second feedback system. Thus, the microphone 111 and the speaker 116 may be connected to the signal processing apparatus 100 of the first feedback system or the signal processing apparatus 100 of the second feedback system.

In fig. 9, the feedforward signal processing apparatus 100 and the signal processing apparatus 100 of the third feedback system are daisy-chained. In this example, a plurality of modules 510, 520, and 530 are connected to form a multiple-input multiple-output signal processing system. In fig. 9, a speaker 116 is connected to the signal processing apparatus 100 of the first feedback system. The cancellation signal is exchanged via communications over the dedicated bus 150. Thus, a speaker may be connected to any signal processing apparatus 100.

Fig. 7 to 9 are merely examples of the connection of the noise canceling system. The combination of the connections is not limited to these. The combination and number of noise canceling systems to be connected may be determined according to the magnitude of noise, the arrival direction of noise, and the like.

The noise is not always uniformly distributed in the processing range where the noise cancellation processing is to be performed. For example, as shown in fig. 10, microphones 111a to 111h and speakers 116a to 116h are connected to a plurality of modules, respectively, and arranged in a circle. The arrival direction of the noise from the noise source 1000 to the microphones 111a to 111h and the speakers 116a to 116h may be concentrated in a specific direction. Therefore, the signal processing apparatuses 100 of the plurality of noise canceling systems are connected in a range where higher-performance noise canceling processing is required, the range being closer to the noise source 1000. The plurality of signal processing apparatuses 100 of the plurality of noise canceling systems as described in fig. 7 to 9 are connected for the arrival direction of noise. Therefore, high-performance noise canceling processing can be performed. Incidentally, for convenience of explanation in fig. 10, only a microphone and a speaker connected to the module will be described.

Further, fig. 11 is an example in which the signal processing apparatus 100 is used as an 8-shaped loop canceller connected to the first feedback system. International publication No. WO 2017/175448 discloses a 8-shaped loop canceller. The signal processing apparatus 100 is applied to a 8-shaped loop canceller. As a result, mutual interference between modules can be reduced.

Further, as shown in fig. 12, a noise analyzer apparatus 600 may be connected to the signal processing apparatus 100. In fig. 12, the audio signal collected by the microphone 111 is supplied to the noise analyzer apparatus 600, and the noise analyzer apparatus 600 performs a predetermined audio analysis process on the audio signal. Thus, the noise analyzer apparatus 600 obtains analysis information such as a noise type, a noise level, a noise arrival direction, and a noise power spectrum. Then, the noise analyzer apparatus 600 supplies the analysis information to the signal processing apparatus 100 via the dedicated bus 150. Therefore, the signal processing apparatus 100 selects a combination of noise canceling systems based on the analysis information. The signal processing apparatus 100 selects the noise cancellation mode. The combination of noise cancellation systems has been described with reference to fig. 5, 6, 7, 8 and 9.

The selection of the noise cancellation mode is to select an airplane mode, an office mode, an outdoor mode, and the like in the signal processing apparatus 100. In each mode, a digital filter, a filter coefficient, and the like are set in advance so that appropriate noise cancellation can be performed according to the magnitude of noise and the type of noise.

As described above, the noise cancellation processing units are daisy-chained. Therefore, the processing range can be expanded, and the performance of noise cancellation can be improved.

[1-3. data Transmission ]

[1-3-1. first example of circular array ]

Next, a first example of data transmission processing between signal processing apparatuses will be described. As shown in fig. 13, the processing range is a region within a specific closed space. In order to reduce noise within the processing range, a plurality of modules are arranged to surround the processing range. The plurality of modules are arranged in a plurality of circular arrays. In fig. 13, there is a noise source 1000 outside the plurality of circular arrays. In fig. 13, only the microphones 111a to 111h and the speakers 116a to 116h connected to the module are shown for convenience of explanation. The signal processing device 100, DA converter 114, AD converter 113, microphone amplifier 112, power amplifier 115, and the like constituting a module are not shown.

Among the microphones 111a to 111h and the speakers 116a to 116h arranged in a plurality of circular arrays, the speaker 116a, the microphone 111a, the speaker 116e, and the microphone 111e, which are located on a straight line, are described. A speaker 116a and a microphone 111a are connected to the module 1. A microphone 111a and a speaker 116e are connected to the module 2. Further, a speaker 116e and a microphone 111e are connected to the module 3. The module 1 performs noise cancellation of the feedback system. Block 2 performs noise cancellation for the feed forward system. The module 3 performs noise cancellation of the feedback system.

When the noise source 1000 is outside the plurality of circular arrays, the noise goes from the outside to the inside of the plurality of circular arrays. That is, the noise reaches the outside of the plurality of circular arrays earlier than the inside. Further, the noise level collected by the microphone 111a arranged outside is higher than the noise level collected by the microphone 111e arranged inside the plurality of circular arrays. Therefore, in order to perform noise cancellation with high accuracy, the importance of the sound collected by the microphone 111a located at the outermost portion of the plurality of circular arrays is high. As the microphones trend toward the interior of multiple circular arrays, the importance of sound collection by these microphones is low. Therefore, it is preferable to transmit the audio signal acquired by the microphone 111a connected to the module 1 positioned at the outermost portion of the plurality of circular arrays to the modules 2 and 3 positioned inside. That is, it can be transferred from the outside to the inside of the plurality of circular arrays, from the high importance circle to the low importance circle. The noise cancellation process in modules located inside the plurality of circular arrays also uses audio signals acquired by microphones connected to modules located outside the plurality of circular arrays.

Fig. 14 shows an outline of data transmission. Fig. 14 shows the relationship between the microphones and the speakers arranged in the circular array shown in fig. 13 by extracting the modules 1, 2, and 3 arranged in a straight line.

Microphone 111a serves as an error microphone in module 1. Microphone 111a, on the other hand, acts as a feedforward reference microphone in module 2, which may collect noise before it reaches the module. That is, this represents a high importance since module 1 is located externally with respect to module 2 and module 1 is close to noise source 1000. Thus, the audio signal is transmitted from the external module 1 to the module 2. Audio signals collected at locations close to the noise source 1000 may be used for noise cancellation processing in modules far from the noise source. The noise cancellation effect can be improved.

Similarly, in module 2 and module 3, module 2 is closer to the noise source 1000. Therefore, the importance of the audio signal acquired by the microphone 111a connected to the module 2 is high. Thus, the audio signal is transmitted from module 1 to module 2. Furthermore, the audio signal is transmitted from module 2 to module 3. Therefore, the audio signal collected at a position close to the noise source 1000 can be used in the noise cancellation process in the module far from the noise source. The noise cancellation effect can be improved.

In addition, the audio signal is transmitted from a module close to the noise source 1000 to a module far from the noise source 1000. In this case, it is preferable to transmit the audio signal by reducing the sampling frequency, reducing the bit rate, or the like. As a result, the data size of the audio signal is reduced. The resources of the signal processing device 100 can be secured, and the transmission speed can be increased.

The present technique increases the number of daisy-chained signal processing devices. Therefore, the scale of the signal processing system can be increased according to the size of the processing range. As the scale of signal processing systems increases, the transmitted data also increases and the size of the processed data also increases. Therefore, it is important to protect resources by reducing the data size in this way. The transmission from a module close to the noise source to a module far from the noise source is a transmission from a module with a high importance level to a module with a low importance level. Thus, the sampling frequency and bit rate are reduced. This does not affect the quality of the noise cancellation even if the quality of the audio signal deteriorates.

[1-3-2. second example of circular array ]

Next, a second example of data transmission processing between the signal processing apparatuses will be described. As shown in fig. 15, the processing range is a region within a specific closed space. In order to reduce noise within the processing range, a plurality of modules are arranged to surround the processing range. The plurality of modules is arranged in a plurality of circular arrays. In fig. 15, a noise source 1000 exists outside the processing range. In fig. 15, only the microphones 111a to 111h and the speakers 116a to 116h connected to the module are shown for convenience of explanation. The signal processing device 100, DA converter 114, AD converter 113, microphone amplifier 112, power amplifier 115, and the like constituting a module are not shown.

Among the microphones 111a to 111h and the speakers 116a to 116h arranged in a plurality of circular arrays, the speaker 116a, the microphone 111a, the speaker 116e, and the microphone 111e, which are located on a straight line, are described. A speaker 116a and a microphone 111a are connected to the module 1. The microphone 111e and speaker 116a are connected to the module 2. Further, a speaker 116e and a microphone 111e are connected to the module 3. The module 1 performs noise cancellation of the feedback system. Block 2 performs noise cancellation for the feed forward system. The module 3 performs noise cancellation of the feedback system.

When the noise source 1000 is inside the plurality of circular arrays, the noise goes from the inside to the outside of the plurality of circular arrays. That is, the noise reaches the inside of the plurality of circular arrays earlier than the outside. Further, the noise level collected by the microphone 111e arranged inside is higher than the noise level collected by the microphones 111a arranged outside the plurality of circular arrays. Therefore, in order to perform noise cancellation with high accuracy, the importance of the sound collected by the microphone 111e positioned at the innermost of the plurality of circular arrays is high. As the microphones go outside of the circular arrays, the sound collected by these microphones is of low importance. Therefore, it is preferable to transmit the audio signal acquired by the microphone connected to the module located innermost among the plurality of circular arrays to the external module. That is, audio signals can be transmitted from the inside to the outside of the plurality of circular arrays, from the high importance circle to the low importance circle. The noise cancellation process in modules located outside the plurality of circular arrays also uses audio signals acquired by microphones connected to modules located inside the plurality of circular arrays.

Fig. 16 shows an outline of data transmission. Fig. 16 shows the relationship between the microphones and the speakers arranged in the circular array shown in fig. 15 by extracting the modules 1, 2, and 3 arranged in a straight line.

The microphone 111e serves as an error microphone in the module 3. On the other hand, the microphone 111e is used as a feedforward reference microphone in the module 2. That is, this represents a high importance since module 3 is located inside with respect to module 2 and close to the noise source 1000. Thus, the audio signal is transmitted from the internal module 3 to the module 2. Audio signals collected at locations close to the noise source 1000 may be used for noise cancellation processing in modules far from the noise source. The noise cancellation effect can be improved.

Similarly, in module 2 and module 1, module 2 is closer to the noise source 1000. Therefore, the importance of the audio signal acquired by the microphone 111e connected to the module 2 is high. Thus, the audio signal is transmitted from module 3 to module 2. Furthermore, the audio signal is transmitted from module 2 to module 1. Therefore, the audio signal collected at a position close to the noise source 1000 can be used in the noise cancellation process in the module far from the noise source. The noise cancellation effect can be improved.

In addition, the audio signal is transmitted from a module close to the noise source 1000 to a module far from the noise source 1000. In this case, it is preferable to transmit the audio signal by reducing the sampling frequency, reducing the bit rate, or the like. This is the same as the example of fig. 11.

The transmission of the audio signal may be similarly performed for a module connected to other microphones and speakers than the microphone 111a, the speaker 116a, the microphone 111e, and the speaker 116e as shown in fig. 13 and 15.

Fig. 17 is a table showing a format of data transmitted between the daisy-chain connected signal processing apparatuses. Although the audio signals are transmitted between the modules in the above description with reference to fig. 13 to 16, the data to be transmitted is not limited to the audio signals. Data transmitted between signal processing devices may be classified into a stream type and a bus type. The data in streaming format includes an audio signal (input of a microphone), a cancellation signal (output of a speaker), a transfer function, and the like. The data in streaming format is required to be real-time.

On the other hand, data in the bus system is control information or the like transmitted and received between connected signal processing apparatuses, the data does not need to have a real-time characteristic, and can be classified into control data and data transmitted and received between modules. The control data is data such as a switching control signal of the noise cancellation process. The data transmitted and received between the modules includes arrangement setting information of the modules, importance information according to the arrangement of the modules, and module numbers. The control data and the data transmitted and received between the modules correspond to control information in the claims.

As specific examples, information indicating the arrival direction of noise, information indicating modules to be connected when a combination of different noise canceling systems is used, information indicating the arrangement relationship of the modules, and the like need not be in real time. It is therefore sufficient to transmit this information in the bus system.

[1-3-3. direction of data Transmission ]

Next, the direction of data transmission will be described with reference to fig. 18. As shown in fig. 18, a plurality of microphones 111a to 111m and a plurality of speakers 116a to 116m connected to a module (not shown) are arranged in a circle. These modules are connected by a dedicated bus.

Microphone 111a, microphone 111b, microphone 111c, speaker 116a, speaker 116b, and speaker 116c are connected to module 1. Microphone 111d, microphone 111e, microphone 111f, speaker 116d, speaker 116e, and speaker 116f are connected to module 2. Microphone 111g, microphone 111h, microphone 111i, speaker 116g, speaker 116h, and speaker 116i are connected to module 3. Microphone 111j, microphone 111k, microphone 111m, speaker 116j, speaker 116k, and speaker 116m are connected to module 4.

In this state, if data transfer is performed only in one of the clockwise direction and the counterclockwise direction, data transfer cannot be efficiently performed. Thus, data is transmitted on the bidirectional bus in both the clockwise and counterclockwise directions. In this way, data in a streaming system requiring real-time performance can be transmitted with low delay.

For example, if data can only be transmitted clockwise and data needs to be transmitted from module 1 to module 2, there is a delay in the transmission. On the other hand, when data transmission can be performed only in the counterclockwise direction, there is a delay in transmission when data transmission from the module 1 to the module 4 is required. Thereby, bidirectional transmission is realized through the bidirectional dedicated bus. Therefore, when data is transmitted from module 1 to module 2, the data can be transmitted with low delay. Furthermore, data may be transmitted with low latency when transmitted from module 1 to module 4.

[1-3-4. packet in data Transmission ]

Next, a first example of a packet in data transmission will be described with reference to fig. 19. The packet in data transmission is a process of collecting data to be transmitted when data is transmitted between modules.

As shown in fig. 19, data is transmitted from module 1 to module 2. In addition, data is transmitted from module 2 to module 3. Note that the numerical values assigned to the data in fig. 19 and fig. 20 described later represent the data in each module. The data transferred from module 1 to module 2 comprises the data in module 1. The data transferred from module 2 to module 3 includes data in module 1 and data in module 2.

Module 1 transmits the data to module 2. Then, the module 2 pulls up the transmitted data once. Next, the acquired data is right shifted. In this way, resources are allocated so that data for module 2 can be inserted. The data of module 2 is then inserted at the beginning of the stream. In this way, data may be transferred from module to module.

Next, a second example of a packet in data transmission will be described with reference to fig. 20. In the first example described above, the shift processing is performed only on the transferred data, and the data size is not changed. On the other hand, data flows tend to be resource limited. For this reason, it may be necessary to reduce the size of data to be transmitted and to reduce the amount of information to be transmitted.

Therefore, as shown in fig. 20, the data size is reduced and transmission is performed. Data is transferred from module 1 to module 2. In addition, data is transmitted from module 2 to module 3. When data is transferred from module 1 to module 2, module 2 first pulls up the transferred data. Second, when the acquired data is an audio signal, the data size is reduced. The data size is reduced by lowering the sampling frequency, lowering the bit rate, and the like. Next, right shifting is performed on the data having the reduced data size. In this way, resources are allocated so that data owned by module 2 itself can be inserted. Then, the data of the module 2 itself is inserted at the beginning of the stream. In this way, the data size is reduced to ensure resources. In addition, data may be transferred from module to module.

In addition, data "1" in fig. 20 indicates 16-bit data of the block 1, and data "1'" indicates 8-bit data of the block 1.

Next, data transmission between modules when the modules 1 to 4 are configured as shown in fig. 18 will be described with reference to fig. 21. In the configuration shown in fig. 18, module 1 and module 2 are adjacent to each other. It can be seen that the contribution rate is high in forming a 3-input 3-output system. In fig. 21, the numerical value assigned to the data represents the data in each module. The data transferred from module 1 to module 2 includes data in module 1, data in module 4, and data in module 3.

Thus, it is assumed that the dedicated bus has bidirectional communication during data transfer. In the data transmission from module 1 to module 2, the audio signal collected by the reference microphone in module 1 is of the highest importance. On the other hand, in the data transmission from module 2 to module 1, the audio signal collected by the reference microphone in module 2 has the highest importance.

Therefore, as shown in fig. 21, data is transmitted from the module 1 to the module 2 in such a manner that the data size of the audio signal collected by the microphone connected to the module 1 becomes the largest. On the other hand, data is transmitted from the module 2 to the module 1 in such a manner that the data size of the audio signal collected by the module 2 becomes the largest. In fig. 21, the data transmitted from module 1 to module 2 includes data that has been transmitted to module 3 and module 4 of module 1. Module 1 is transferred to module 2 with the largest data size. The data transmitted from module 2 to module 1 comprises data that has been transmitted to modules 3 and 4 of module 2. The data of module 2 is transferred to module 1 having the largest size.

Fig. 22 is a block diagram showing a process when a multiple-input multiple-output process is performed using audio signals (hereinafter, referred to as reference signals) collected by reference microphones of two adjacent modules. Fig. 22 generally illustrates noise cancellation for a feed forward system in determinant.

Here, in the module 1 and the module 2 shown in fig. 18, a reference signal collected by a microphone connected to the module 2 is transmitted to the module 1, and the reference signal is used in the module 1. In this case, what signal processing is possible in the module 1 will be described. Fig. 23 shows a determinant in the module 1. Module 1 may use the audio signal collected by the reference microphone of module 1 and the audio signal collected by the reference microphone of module 2 as reference signals. Therefore, the calculation amount corresponding to the portion surrounded by the broken line of the control filter H in fig. 23 can be allocated to the module 1. The module 2 is similar.

Fig. 24 shows a signal processing block diagram of a second feedback system in a multiple-input multiple-output system. The basic configuration is the same process as that shown in fig. 21. In the second feedback system of the mimo system, the output signal represented by the thick line also needs to be transmitted in the same manner as the error signal.

The signal processing apparatus according to the present technology is configured as described above. According to the present technology, the scale of a signal processing system that performs noise cancellation can be easily extended by daisy-chain connection. For example, as the multiple-input multiple-output processing, a multiple-input multiple-output feedforward noise canceling process and a multiple-input multiple-output feedback process may be performed.

Further, the multiple-input multiple-output system can be controlled by communication using a dedicated bus. This makes it possible to use a module suitable for controlling the scale. For example, a system using both a feed-forward system and a feedback system is implemented, and in addition, two feedback systems are used together. Algorithms with high noise reduction performance may be employed.

Control information between connected signal processing apparatuses can be managed by communication using a dedicated bus. An appropriate filter for noise cancellation may be selected. Noise cancellation may be turned on or off.

Further, the signal processing system is configured in a circular form. Therefore, it is possible to effectively reduce noise reaching the inside from the outside by performing processing in multiple stages. Alternatively, noise that reaches the outside from the inside can be effectively reduced by performing processing in multiple stages. When the signal processing system is configured in a circular shape, data is transmitted according to the importance level. Thus, the audio signal collected by the microphone in the outer circle is used as an error microphone for the external speaker. The audio signal is used as a reference microphone for the internal speaker. This can improve the noise cancellation performance.

<2. modification >

The embodiments of the present technology have been described above in detail, but the present technology is not limited to the above-described embodiments, and various modifications based on the technical idea of the present technology are possible.

The connection of the plurality of signal processing apparatuses 100 is not limited to the dedicated bus 150. If the effects of the present technology can be achieved, a plurality of signal processing apparatuses 100 may be connected by a general-purpose bus. Further, the connection of the plurality of signal processing apparatuses 100 is not limited to the daisy chain connection. The plurality of signal processing apparatuses 100 may be connected in other connection forms as long as the effects of the present technology can be achieved. Other forms of connection include, for example, star, ring, etc.

The present technology can also be configured as follows.

(1) A signal processing apparatus comprising:

a noise removal processing unit connectable to the one or more input units and connectable to the one or more output units, the plurality of signal processing devices being connected to each other and configured to perform noise removal processing.

(2) The signal processing apparatus according to item (1), wherein

The plurality of signal processing devices are daisy-chained.

(3) The signal processing device according to the item (1) or the item (2), wherein

Data is transmitted between the plurality of noise canceling processing units.

(4) The signal processing apparatus according to item (3), wherein

The data is an audio signal input from one or more input units.

(5) The signal processing device according to the item (3) or the item (4), wherein

The data is a cancellation signal output from one or more output units.

(6) The signal processing device according to the item (3) or the item (4), wherein

The data is control information.

(7) The information processing apparatus according to any one of the items (3) to (6), wherein

The size of the data is reduced and data transmission is performed.

(8) The information processing apparatus according to item (7), wherein,

the size of the data is reduced by reducing the sampling frequency.

(9) The information processing apparatus according to item (7) or item (8), wherein

The size of the data is reduced by reducing the bit rate.

(10) The signal processing apparatus according to any one of the items (1) to (9), wherein

One of the plurality of noise canceling processing units to which the input unit close to the noise source is connected transmits data to another one of the plurality of noise canceling processing units to which the input unit far from the noise source is connected.

(11) The signal processing apparatus according to any one of the items (1) to (10), wherein

The signal processing device is connected to a noise analyzer unit that analyzes noise, and switches noise cancellation processing according to an analysis result of the noise analyzer unit.

(12) The signal processing apparatus according to the item (11), wherein

The signal processing device switches the mode of the noise cancellation process according to the analysis result of the noise analyzer unit.

(13) The signal processing device according to the item (11) or the item (12), wherein

The noise removal processing unit is capable of performing noise removal processing of a plurality of systems and changing a combination of the systems according to an analysis result of the noise analyzer unit.

(14) The signal processing apparatus according to any one of the items (1) to (13), wherein

The plurality of input units and the plurality of output units are connected to any one of a plurality of noise cancellation processing units connected to each other.

(15) A signal processing method, comprising:

a plurality of signal processing devices are connected to each other and perform noise cancellation processing, each of the plurality of signal processing devices including a noise cancellation processing unit connectable to one or more input units and connectable to one or more output units.

(16) A signal processing program that causes a computer to execute a signal processing method, the signal processing method comprising:

a plurality of signal processing devices are connected to each other and perform noise cancellation processing, each of the plurality of signal processing devices including a noise cancellation processing unit connectable to one or more input units and connectable to one or more output units.

List of reference numerals

100 signal processing device

101 noise elimination processing unit

111 microphone

116 speaker.