Noise analysis device, noise analysis method, and program

Abstract

To provide a noise analysis device, a noise analysis method, and a program that explicitly notify a user of how much noise that was introduced during execution of specific processing by the user is present.SOLUTION: A telephone call system includes at least a plurality of sound processing devices connected via a network such that they can perform data communication with each other. In the telephone call system, a sound processing device 10 comprises: a sound analysis unit 13 that separates an input sound signal collected by a sound collection device into a noise sound signal and a non-noise sound signal; a sound A sound volume analysis unit 14 that analyzes a sound volume of the non-noise sound signal; a sound B sound volume analysis unit 15 that analyzes the sound volume of the noise signal; and a control unit 17 that displays a first analysis result as an analysis result for the sound volume of the non-noise sound signal, and a second analysis result as an analysis result for the sound volume of the noise sound signal, on a display in a visual and comparative manner.SELECTED DRAWING: Figure 2

Description

[Technical field]

[0001] The present disclosure relates to a noise analysis device, a noise analysis method, and a program. [Background technology]

[0002] Patent document 1 discloses an automatic volume control device that determines the voice and non-voice parts of an input audio signal, calculates the voice level in the voice part, calculates the noise level in the non-voice part, calculates the S / N ratio of the input audio signal from the voice level and noise level, calculates the amplification factor of the input audio signal from the voice level, noise level, S / N ratio and a preset target volume level, and amplifies the input audio signal with the amplification factor and outputs it. [Prior art documents] [Patent documents]

[0003] [Patent Document 1] JP 2009-171208 A Summary of the Invention [Problem to be solved by the invention]

[0004] In recent years, remote calls have been increasing, and opportunities to make remote calls even while working from home are increasing. However, when making a remote call while working from home, noise such as noise around the speaker is often mixed into the call voice, and in that case, noise is also sent to the other party, which may cause a problem in conversation or conference. In addition, there is no effective means for the speaker of the remote call to clearly know how much mixed noise is being transmitted to the other party. Patent Document 1 can suppress stationary noise, but sudden noise cannot be suppressed and is transmitted to the other party. In addition, even with recent noise removal using artificial intelligence, it is difficult for users to grasp what type of noise can be removed. For example, if a baby is crying near the speaker of the remote call, it is not possible to grasp how much noise the baby's crying is being transmitted to the other party of the remote call. Patent Document 1 also does not disclose an effective means for the speaker to clearly know how much noise such as the mixed noise is being transmitted to the other party of the remote call.

[0005] The present disclosure has been devised in consideration of the above-described conventional circumstances, and aims to provide a noise analysis device, a noise analysis method, and a program that explicitly notify a user of the level of noise that has been mixed in while the user is performing a specific process. [Means for solving the problem]

[0006] The present disclosure provides a noise analysis device including a sound analysis unit that separates an input sound signal collected by a sound collection device into a noise sound signal and a non-noise sound signal, a first analysis unit that analyzes the volume of the non-noise sound signal, a second analysis unit that analyzes the volume of the noise sound signal, and a display control unit that visually and contrastively displays on a display the first analysis result that is an analysis result of the volume of the non-noise sound signal and the second analysis result that is an analysis result of the volume of the noise sound signal.

[0007] The present disclosure also provides a noise analysis method executed by a noise analysis device, the noise analysis method comprising the steps of: separating an input sound signal collected by a sound collection device into a noise sound signal and a non-noise sound signal; analyzing the volume of the non-noise sound signal; analyzing the volume of the noise sound signal; and visually and contrastively displaying on a display a first analysis result that is an analysis result of the volume of the non-noise sound signal and a second analysis result that is an analysis result of the volume of the noise sound signal.

[0008] The present disclosure also provides a program for causing a noise analysis device, which is a computer, to realize the steps of separating an input sound signal collected by a sound collection device into a noise sound signal and a non-noise sound signal, analyzing the volume of the non-noise sound signal, analyzing the volume of the noise sound signal, and visually and contrastively displaying on a display a first analysis result that is the analysis result of the volume of the non-noise sound signal and a second analysis result that is the analysis result of the volume of the noise sound signal. Effect of the Invention

[0009] According to the present disclosure, it is possible to explicitly notify the user of the level of noise present during execution of a specific process. [Brief description of the drawings]

[0010] [Figure 1] FIG. 1 is a diagram showing an example of a system configuration of a call system according to a first embodiment; [Diagram 2] FIG. 2 is a block diagram showing an example of the hardware configuration of the sound processing device shown in FIG. [Figure 3A] FIG. 11 shows a first example of an audio meter and a noise meter displayed in a display area. [Figure 3B] FIG. 2 shows a second example of the display of the audio meter and noise meter displayed in the display area. [Figure 3C] FIG. 3 shows a third example of the display of the audio meter and noise meter displayed in the display area. [Figure 4]1 is a flowchart showing an example of an operation procedure of the sound processing device according to the first embodiment in a chronological order; [Diagram 5] 4. A flowchart showing a detailed operation procedure example of the sound analysis in step St2 of FIG. 4 in chronological order. [Figure 6A] 5 is a flowchart showing a detailed first example of an operation procedure of the volume comparison and correction process in step St5 of FIG. 4 in chronological order. [Figure 6B] 5 is a flowchart showing a detailed second operation procedure example of the volume comparison and correction process in step St5 of FIG. 4 in chronological order. [Figure 7] 5 is a flowchart showing a detailed operation procedure example of the volume display process in step St6 of FIG. 4 in chronological order. [Figure 8] 5 is a flowchart showing a detailed operation procedure example of signal processing parameter calculation in step St7 of FIG. 4 in chronological order. [Figure 9] FIG. 13 is a diagram showing an example of a system configuration of a voiceprint authentication system according to a second embodiment. [Figure 10] FIG. 10 is a block diagram showing an example of a hardware configuration of the sound processing device shown in FIG. [Figure 11] A flowchart showing a detailed operation procedure example of voiceprint authentication parameter calculation in step St7 of FIG. 4 according to the second embodiment in a chronological order. [Figure 12] FIG. 1 is a block diagram showing an example of a hardware configuration of a sound processing device according to a first modification of the first embodiment. [Figure 13] 1 is a flowchart showing an example of an operation procedure of a sound processing device according to a first modification of the first embodiment in a chronological order. [Figure 14] FIG. 11 is a block diagram showing an example of a hardware configuration of a sound processing device according to a second modification of the first embodiment. [Figure 15] 1 is a flowchart showing an example of an operation procedure of a sound processing device according to a second modification of the first embodiment in a chronological order. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0011] Hereinafter, with reference to the drawings as appropriate, a detailed description will be given of an embodiment specifically disclosing a noise analysis device, a noise analysis method, and a program according to the present disclosure. However, more detailed description than necessary may be omitted. For example, detailed description of already well-known matters or duplicate description of substantially the same configuration may be omitted. This is to avoid the following description becoming unnecessarily redundant and to facilitate understanding by those skilled in the art. Note that the attached drawings and the following description are provided to enable those skilled in the art to fully understand the present disclosure, and are not intended to limit the subject matter described in the claims.

[0012] <Embodiment 1> In embodiment 1, as one implementation of a noise analysis device according to the present disclosure, a call system that supports remote calls between multiple people using a sound processing device and a headset connected to each other via a network for data communication is exemplified (see FIG. 1).

[0013] First, with reference to Fig. 1, a system configuration example of a call system 100 according to the first embodiment will be described. Fig. 1 is a diagram showing a system configuration example of the call system 100 according to the first embodiment. The call system 100 includes at least sound processing devices 10, 20, 30, and 40 connected to each other so as to be able to perform data communication via a network NW1. Headsets HS1, HS2, HS3, and HS4 for remote calls and displays DP1, DP2, DP3, and DP4 for display are connected to the sound processing devices 10, 20, 30, and 40, respectively. The headsets HS1, HS2, HS3, and HS4 are used by users SP1, SP2, SP3, and SP4, respectively.

[0014] The network NW1 may be a wired network, a wireless network, or a combination of a wired network and a wireless network. The wired network may be, for example, a wired Local Area Network (LAN) or a wired Wide Area Network (WAN), etc. The wireless network may be, for example, a wireless LAN, a wireless WAN, or a cellular wireless communication, etc.

[0015] For simplicity's sake, the following description focuses on the configuration and operation of a sound processing device 10 that visually and comparatively displays the volume of the speech of user SP1 picked up by a headset HS1 used by user SP1 and the volume of noise around user SP1, with a focus on user SP1 who is the speaker of a remote call. When user SP1 is the speaker of a remote call, other users SP2, SP3, and SP4 are each call partners.

[0016] The sound processing device 10 is an example of a noise analysis device according to the present disclosure, and visually contrasts the volume of the speech of the user SP1 with the volume of the surrounding noise in a display area AR1 that is a part of the display DP1 (see Figs. 3A, 3B, and 3C). Note that the description of the configuration and operation of the sound processing device 10 may be similarly applied to some or all of the other sound processing devices 20, 30, and 40.

[0017] The headset HS1 may have the same configuration as a known headset, and includes a microphone and a speaker. The headset HS1 picks up the speech of the user SP1 during a remote call with a microphone, and outputs the speech of at least one of the call partners, users SP2, SP3, and SP4, from the speaker. The sound picked up by the microphone of the headset HS1 includes not only the speech of the user SP1, but also noise such as life noise present around the user SP1. The life noise here includes, for example, the crying of a baby around the user SP1, sounds that occur suddenly in the surroundings such as the ringing of an intercom or a ringtone of a telephone, and low-volume noise that occurs constantly.

[0018] The display DP1 is, for example, a liquid crystal display or an organic electroluminescence (EL) display, which receives a display data signal from the sound processing device 10 and displays an image or the like corresponding to the data signal. The display DP1 visually and contrastively displays, for example, the volume of the user SP1's speech and the volume of surrounding noise within a display area AR1 (see Figs. 3A, 3B, and 3C). The display area AR1 is a rectangular area at the lower right edge of the entire image / video display area of ​​the display DP1, as an example of a position that is easily noticeable to the user SP1. However, the position of the display area AR1 does not have to be limited to the lower right edge as described above, as long as it is a position that is easily noticeable to the user SP1.

[0019] Next, with reference to FIG. 2, an example of a hardware configuration of the sound processing device 10 of FIG. 1 will be described. FIG. 2 is a block diagram showing an example of a hardware configuration of the sound processing device 10 of FIG. 1. The sound processing device 10 is connected to a headset HS1 and a display DP1 so as to be capable of inputting and outputting data signals. The sound processing device 10 is a general-purpose computer device, and may be, for example, a personal computer, a smartphone, a tablet terminal, or a server computer. The sound processing device 10 includes at least a processor PRC1 and a memory 11 as hardware components. The sound processing device 10 may further include an input device such as a mouse that accepts operations by a user SP1.

[0020] The processor PRC1 functions as a controller that manages the overall operation of the sound processing device 10, and performs control processing for managing the operations of each part of the sound processing device 10, input / output processing of data signals / control signals between each part of the sound processing device 10, arithmetic processing of data signals / control signals, and storage processing of data signals / control signals. The processor PRC1 may be configured, for example, with a Central Processing Unit (CPU), a Digital Signal Processor (DSP), a Field Programmable Gate Array (FPGA), or a Graphical Processing Unit (GPU). The processor PRC1 inputs a sound signal (hereinafter, sometimes referred to as an "input sound signal" for convenience) collected by the headset HS1 in cooperation with the memory 11, and performs various signal processing on the input sound signal.

[0021] The memory 11 includes at least a Read Only Memory (ROM) and a Random Access Memory (RAM). The ROM stores programs that define various processes executed by the processor PRC1 and data required for executing the programs in association with each other. The RAM temporarily stores data or information acquired or generated by the processor PRC1. The memory 11 is referred to as necessary during the execution of processes by the processor PRC1.

[0022] Here, a detailed description will be given of an example of the configuration of the executing entity of the processing functionally executed by the processor PRC1. The processor PRC1 functionally includes a sound signal input IF 12, a sound analysis unit 13, a sound A volume analysis unit 14, a sound B volume analysis unit 15, a volume comparison and correction unit 16, a control unit 17, a signal processing unit 18, and a transmission unit 19.

[0023] The sound signal input IF12 has a function as an interface (IF) with the headset HS1, and receives an input sound signal picked up by the headset HS1 and sends it to each of the sound analysis unit 13 and the signal processing unit 18. In the following description, while the power of the sound processing device 10 is ON, the input sound signal picked up by the headset HS1 continues to be input to the sound signal input IF12 periodically, intermittently, or steadily.

[0024] The sound analysis unit 13 separates the input sound signal from the sound signal input IF 12 into at least two types of sound signals (specifically, sound A and sound B), sends the sound A signal (an example of a non-noise sound signal) to the sound A volume analysis unit 14, and sends the sound B signal (an example of a noise sound signal) to the sound B volume analysis unit 15. Here, sound A refers to the sound of the part of the speech voice uttered by the user SP1 among the sounds contained in the input sound signal. In other words, sound A refers to the sound of the call voice that is assumed to be delivered to the call partners, users SP2, SP3, and SP4, by the speech of the user SP1 who is the speaker of the remote call. On the other hand, sound B refers to noise sound other than the speech voice uttered by the user SP1 among the sounds contained in the input sound signal. In other words, sound B is an unnecessary noise sound mixed with sound A delivered to users SP2, SP3, and SP4, and is a sound that may be an obstacle when users SP2, SP3, and SP4 hear the speech voice of user SP1. The processing by the sound analysis unit 13 will be described in detail later with reference to FIG.

[0025] The sound A volume analysis unit 14 is an example of a first analysis unit, and inputs the sound A signal from the sound analysis unit 13 and analyzes the volume of the input sound A signal. The sound A volume analysis unit 14 analyzes the volume of the sound A signal by referring to, for example, the average power, maximum amplitude, or power in a specific frequency band of the waveform of the sound A signal. The sound A volume analysis unit 14 may also smooth the waveform of the input sound A signal in the time direction, and then analyze the volume of the sound A signal by referring to the average power, maximum amplitude, or power in a specific frequency band of the waveform of the sound A signal. The sound A volume analysis unit 14 sends the analysis result of the volume of the sound A signal (first analysis result) to the volume comparison and correction unit 16.

[0026] The sound B volume analysis unit 15 is an example of a second analysis unit, and inputs the sound B signal from the sound analysis unit 13 and analyzes the volume of the input sound B signal. The sound B volume analysis unit 15 analyzes the volume of the sound B signal, for example, by referring to the average power, maximum amplitude, or power in a specific frequency band of the waveform of the sound B signal. The sound B volume analysis unit 15 may also smooth the waveform of the input sound B signal in the time direction, and then analyze the volume of the sound B signal by referring to the average power, maximum amplitude, or power in a specific frequency band of the waveform of the sound B signal. The sound B volume analysis unit 15 sends the analysis result of the volume of the sound B signal (second analysis result) to the volume comparison and correction unit 16.

[0027] The volume comparison correction unit 16 inputs and compares the first analysis result from the sound A volume analysis unit 14 and the second analysis result from the sound B volume analysis unit 15. When the first analysis result and the second analysis result satisfy a specific first condition, the volume comparison correction unit 16 may correct the volume of the sound A signal to be smaller than the value of the volume analyzed by the sound A volume analysis unit 14 (see FIG. 6A and FIG. 6B), and in this case notifies the signal processing unit 18 of the corrected volume of the sound A signal. As a result of the comparison, the volume comparison correction unit 16 generates a display data signal for visually and comparatively displaying the respective volumes of the sound A signal and the sound B signal (see FIG. 7), and visually and comparatively displays the respective volumes of the sound A signal and the sound B signal in the display area AR1 of the display DP1. In addition, the volume comparison correction unit 16 sends the second analysis result to the control unit 17 every time it obtains the second analysis result. Details of the processing of the volume comparison correction unit 16 will be described later with reference to FIG. 6A, FIG. 6B, and FIG. 7, respectively.

[0028] The control unit 17 judges whether or not the second analysis result input from the volume comparison and correction unit 16 satisfies a specific second condition (see FIG. 8). When the control unit 17 judges that the second analysis result satisfies the specific second condition, it generates a data signal of a predetermined message (see below) and displays it in an area other than the display area AR1 within the entire image / video display area of ​​the display DP1. Details of the processing of the control unit 17 will be described later with reference to FIG. 8.

[0029] The signal processing unit 18 performs predetermined signal processing on the input sound signal from the sound signal input IF 12 and sends the signal to the transmission unit 19. When the signal processing unit 18 receives notification of the corrected volume of the sound A signal from the volume comparison and correction unit 16, the signal processing unit 18 attenuates the volume of the input sound signal from the sound signal input IF 12 to the corrected volume of the sound A signal, and sends the signal to the transmission unit 19.

[0030] The transmitting unit 19 transmits the input sound signal that has not been attenuated or the input sound signal that has been attenuated from the signal processing unit 18 as a call voice signal mixed with noise sound via the network NW1 to each of the other sound processing devices 20, 30, 40. Note that although the transmitting unit 19 is illustrated as constituting a part of the processor PRC1, it may be provided separately from the processor PRC1.

[0031] Next, display examples of the audio meter and noise meter displayed in the display area AR1 will be described with reference to Fig. 3A to Fig. 3C. Fig. 3A is a diagram showing a first display example of the audio meter and noise meter displayed in the display area. Fig. 3B is a diagram showing a second display example of the audio meter and noise meter displayed in the display area. Fig. 3C is a diagram showing a third display example of the audio meter and noise meter displayed in the display area.

[0032] As shown in FIG. 3A, the display area AR1 visually and contrastively displays an audio meter V1 corresponding to a first analysis result, which is an analysis result of the volume of the sound A signal, and a noise meter N1 corresponding to a second analysis result, which is an analysis result of the volume of the sound B signal. The audio meter V1 is displayed in one display frame Wk1, and indicates the volume level by the number of scales (display bars) corresponding to the volume level of the sound A signal. Similarly, the noise meter N1 is displayed in one display frame Wk2, and indicates the volume level by the number of scales (display bars) corresponding to the volume level of the sound B signal. In the display example of FIG. 3A, since the scales (display bars) of the audio meter V1 are more than the scales (display bars) of the noise meter N1, it can be seen that the voice of the user SP1, who is the speaker, is collected more than the noise sound. In addition, the user SP1 can visually and contrastively recognize that the noise sound indicated by the scale of the noise meter N1 is mixed in with the volume of his / her own voice (i.e., the scale of the audio meter V1).

[0033] As shown in FIG. 3B, in the display area AR1, a voice meter V2 corresponding to a first analysis result, which is an analysis result of the volume of the sound A signal, and a noise meter N2 corresponding to a second analysis result, which is an analysis result of the volume of the sound B signal, are visually and contrastively displayed in one display frame Wk3. The voice meter V2 is displayed in the same display frame Wk3, and indicates the volume by the number of scales (display bars) corresponding to the volume of the sound A signal. Similarly, the noise meter N2 is displayed in the same display frame Wk3, and indicates the volume by the number of scales (display bars) corresponding to the volume of the sound B signal. In the display example of FIG. 3B, since the scales (display bars) of the voice meter V2 are fewer than the scales (display bars) of the noise meter N2, it can be seen that the surrounding noise sounds are more prominent than the voice of the user SP1 who is the speaker. In addition, the user SP1 can visually and contrastively recognize that there is a lot of noise sound indicated by the scale of the noise meter N2 compared to the volume of his / her own voice (i.e., the scale of the voice meter V2).

[0034] As shown in FIG. 3C, the display area AR1 visually and contrastively displays a voice meter V3 corresponding to the first analysis result, which is the analysis result of the volume of the sound A signal, and a noise meter N3 corresponding to the second analysis result, which is the analysis result of the volume of the sound B signal, in a pie chart format. The zero (origin) of each of the voice meter V3 and the noise meter N3 is defined by the dotted line I1. That is, with the position of the dotted line I1 as the zero point, the volume of the sound A signal is indicated in a pie chart format by the voice meter V3, and the volume of the sound B signal is indicated in a pie chart format by the noise meter N3. In the display example of FIG. 3C, since the arc length (length of the display bar) of the voice meter V3 is longer than the arc length (length of the display bar) of the noise meter N3, it can be seen that the voice of the user SP1, who is the speaker, is collected in preference to the noise sound. In addition, the user SP1 can visually and contrastively recognize that the noise sound indicated by the arc length (length of the display bar) of the noise meter N3 is mixed in with the volume of his / her own voice (i.e., the arc length of the voice meter V3).

[0035] Next, an example of an operation procedure of the sound processing device 10 according to the first embodiment will be described with reference to Figs. 4 to 8. Fig. 4 is a flowchart showing an example of an operation procedure of the sound processing device 10 according to the first embodiment in a chronological order. Fig. 5 is a flowchart showing a detailed example of an operation procedure of the sound analysis of step St2 in Fig. 4 in a chronological order. Fig. 6A is a flowchart showing a detailed first example of an operation procedure of the volume comparison and correction process of step St5 in Fig. 4 in a chronological order. Fig. 6B is a flowchart showing a detailed second example of an operation procedure of the volume comparison and correction process of step St5 in Fig. 4 in a chronological order. Fig. 7 is a flowchart showing a detailed example of an operation procedure of the volume display process of step St6 in Fig. 4 in a chronological order. Fig. 8 is a flowchart showing a detailed example of an operation procedure of the signal processing parameter calculation of step St7 in Fig. 4 in a chronological order.

[0036] First, an example of the overall operation procedure of the sound processing device 10 according to the first embodiment will be described with reference to Fig. 4. Each process shown in Fig. 4 is mainly executed by the processor PRC1 (see Fig. 2).

[0037] 4, the processor PRC1 holds the input sound signal input from the headset HS1 in the sound signal input IF 12 for a certain period of time (step St1). The certain period of time is, for example, 1 second, 5 seconds, 10 seconds, etc. The processor PRC1 performs sound analysis on the input sound signal for the certain period of time held in step St1, and separates the input sound signal into a sound A signal and a sound B signal in the sound analysis unit 13 and outputs them (step St2). A detailed example of the operation procedure of the sound analysis in step St2 will be described later with reference to FIG. 5.

[0038] The processor PRC1 uses the sound A signal obtained in step St2 to analyze the volume of the sound A signal in the sound A volume analysis unit 14 (step St3). The processor PRC1 uses the sound B signal obtained in step St2 to analyze the volume of the sound B signal in the sound B volume analysis unit 15 (step St4). The processor PRC1 inputs the analysis result of the volume of the sound A signal obtained by the analysis in step St3 (first analysis result) and the analysis result of the volume of the sound B signal obtained by the analysis in step St4 (second analysis result) into the volume comparison and correction unit 16 for comparison (step St5). A detailed example of the operation procedure of the volume comparison and correction process in step St5 will be described later with reference to each of Figs. 6A and 6B.

[0039] The processor PRC1 performs a volume display process (see FIG. 7) using the analysis result (first analysis result) of the volume of the sound A signal obtained by the analysis in step St3 or the output after correction in step St5, and the analysis result (second analysis result) of the volume of the sound B signal obtained by the analysis in step St4 or the output after correction in step St5, thereby generating a display data signal for visually and comparatively displaying the respective volumes of the sound A signal and the sound B signal (step St6). A detailed example of the operation procedure of the volume display process in step St6 will be described later with reference to FIG. 7.

[0040] After the volume display process in step St6, the processor PRC1 performs a calculation process of parameters for signal processing (step St7) using the analysis result (second analysis result) of the volume of the sound B signal obtained by the analysis in step St4. A detailed example of the operation procedure of the calculation process in step St7 will be described later with reference to FIG.

[0041] After step St7, if the end of the remote call is instructed by the input operation of the user SP1 (step St8, YES), the process of the processor PRC1 shown in Fig. 4 ends. On the other hand, if the end of the remote call is not instructed by the input operation of the user SP1 (step St8, NO), the process of the processor PRC1 returns to step St1. In other words, the processor PRC1 repeatedly executes a series of processes from step St1 to step St8 until the end of the remote call is instructed by the input operation of the user SP1.

[0042] Next, a detailed example of the operation procedure of the sound analysis in step St2 in Fig. 4 will be described with reference to Fig. 5. Each process shown in Fig. 5 is mainly executed by the sound analysis unit 13 of the processor PRC1 (see Fig. 2).

[0043] 5, the sound analysis unit 13 uses the input sound signal for a certain period held in step St1 to extract one or more feature quantities of the input sound signal (step St2-1). To extract the feature quantities, the sound analysis unit 13 performs, for example, Fast Fourier Transform (FFT) processing or various analyses to obtain Mel Frequency Cepstral Coefficients (MFCCs). Note that the type of analysis performed by the sound analysis unit 13 to extract the feature quantities is not particularly limited.

[0044] The sound analysis unit 13 compares one or more features of the input sound signal extracted in step St2-1 with registered features (e.g., features of sounds registered in advance according to the type of sound A) previously stored in the memory 11, and converts the input sound signal using a filter or the like so that it approaches the registered features, thereby outputting a sound A signal (step St2-2). The sound analysis unit 13 uses one or more features of the input sound signal extracted in step St2-1 and other registered features (e.g., features of sounds registered in advance according to the type of sound B) previously registered in the memory 11, and converts the input sound signal using a filter or the like so that it approaches the other registered features, in the same manner as in step St2-2, thereby outputting a sound B signal (step St2-3). This allows the sound analysis unit 13 to separate and extract the sound A signal and the sound B signal from the input sound signal for a certain period held in step St1.

[0045] In addition, as long as the sound analysis unit 13 can extract the sound A signal and the sound B signal from the input sound signal for a certain period, it is not necessary to use a conversion method using a filter or the like based on a comparison with registered features previously stored in the memory 11. For example, the sound analysis unit 13 may convert and process the input sound signal for a certain period into a sound A signal and also convert and process the input sound signal into a sound B signal according to a predetermined rule. In addition, the sound analysis unit 13 may input the input sound signal for a certain period and output two or more sound signals (for example, a sound A signal and a sound B signal) using a neural network or the like. In addition, in steps St2-2 and St2-3, the sound analysis unit 13 may output the feature amount of each sound signal or a parameter that allows the volume of the sound signal to be referred to, in addition to the mode of outputting the waveform of the sound signal, with respect to the output of the sound A signal and the sound B signal.

[0046] Next, a detailed operation procedure example of the volume comparison and correction process in step St5 in Fig. 6A will be described with reference to Fig. 6A. Each process shown in Fig. 6A is mainly executed by the volume comparison and correction unit 16 of the processor PRC1 (see Fig. 2).

[0047] In FIG. 6A, the volume comparison correction unit 16 compares the analysis result of the volume of the sound A signal (the first analysis result) with the analysis result of the volume of the sound B signal (the second analysis result), and determines whether or not the formula "B ≧ k * A?" (an example of a specific first condition) is satisfied (step St5-1). * is the multiplication operator, and the same applies hereinafter. That is, the volume comparison correction unit 16 determines whether or not "the volume of the sound B signal" is equal to or greater than "k times the volume of the sound A signal" (in other words, whether or not the noise sound is equal to or greater than the target sound, which is the speech of user SP1). k is a constant, for example, k = 1 or k = 0.5. If it is determined that "B < k * A" (step St5-1, NO), the process of the volume comparison correction unit 16 in FIG. 6A ends. That is, in this case, the volume comparison correction unit 16 uses the analysis result of the volume of the sound A signal (the first analysis result) and the analysis result of the volume of the sound B signal (the second analysis result) as they are without correction for the volume display process.

[0048] On the other hand, when the volume comparison correction unit 16 determines that "B ≧ k * A" (step St5-1, YES), it corrects the volume of the sound A signal to be smaller than the volume of the sound A signal obtained by the analysis in step St3 using the coefficient value m (for example, 0.5 or 0.25) (step St5-2). That is, when the volume of the sound A signal obtained by the analysis in step St3 is set as A, the volume comparison correction unit 16 calculates the volume of the sound A signal as m * A in step St5-2. When the surrounding noise sound is equal to or greater than the target sound, which is the speech of user SP1, it is presumed that the target sound is difficult to hear for the other party in the remote call. Accordingly, the volume comparison correction unit 16 can accurately make user SP1 understand the current situation of the remote call of user SP1 by correcting the volume of the sound A signal to be smaller as in step St5-2. After step St5-2, the process of the volume comparison correction unit 16 in FIG. 6A ends.

[0049] Next, with reference to FIG. 6B, a detailed operation procedure example of the volume comparison correction process in step St5 of FIG. 6B will be described. Each process shown in FIG. 6B is mainly executed by the volume comparison correction unit 16 of the processor PRC1 (see FIG. 2).

[0050] 6B, the volume comparison correction unit 16 determines whether or not the analysis result of the volume of the sound A signal (first analysis result) is equal to or greater than a certain volume value (an example of a first threshold value) stored in advance in the memory 11 (step St5-11). If it is determined that the volume of the sound A signal is less than the certain volume value (step St5-11, NO), the processing of the volume comparison correction unit 16 in FIG. 6B ends. That is, in this case, the volume comparison correction unit 16 uses the analysis result of the volume of the sound A signal (first analysis result) and the analysis result of the volume of the sound B signal (second analysis result) as they are in the volume display processing without correcting them.

[0051] On the other hand, if the volume comparison and correction unit 16 determines that the volume of the sound A signal is equal to or greater than a certain volume value (step St5-11, YES), it determines whether or not the sound B signal extracted from the input sound signal for the previous certain period is equal to or greater than a certain volume value (an example of a first threshold value) (step St5-12).If it determines that the volume of the sound B signal extracted from the input sound signal for the previous certain period is less than the certain volume value (step St5-12, NO), the processing of the volume comparison and correction unit 16 in FIG. 6B ends.

[0052] On the other hand, when the volume comparison correction unit 16 determines that the volume of the sound B signal extracted from the input sound signal for the previous certain period is equal to or greater than a certain volume value (step St5-12, YES), it corrects the volume of the sound A signal using a coefficient value n (for example, 0 or 0.01) so that it is smaller than the volume of the sound A signal obtained by the analysis in step St3 (step St5-13). In other words, when the volume of the sound A signal obtained by the analysis in step St3 is A, the volume comparison correction unit 16 calculates the volume of the sound A signal as n*A in step St5-13. The accuracy of the separation into sound A and sound B by the sound analysis unit 13 is not actually perfectly high. In reality, the sound analysis unit 13 may output the sound B signal as the sound A signal as a processing result of the sound analysis. In addition, such output errors tend to be caused by performing the process in a short time, and sound A (or sound B) is often mistaken for sound B (or sound A) when sound A (or sound B) is being output. Based on this, the volume comparison and correction unit 16 can correct or invalidate the instantaneous output values ​​of sounds A and B by the process of Fig. 6B, so as not to inform the user SP1 of the output error. Note that in each process of steps St5-11, St5-12, and St5-13 of Fig. 6B, a process of replacing sound A with sound B, sound B with sound A, and sound A with sound B, respectively, may be performed.

[0053] Next, a detailed operation procedure example of the volume display process in step St6 in Fig. 4 will be described with reference to Fig. 7. The processes shown in Fig. 7 are mainly executed by the volume comparison and correction unit 16 of the processor PRC1 (see Fig. 2).

[0054] In Fig. 7, the volume comparison correction unit 16 converts the levels (volumes) of the sound A signal and the sound B signal into a predetermined logarithm (step St6-1) using the analysis result (first analysis result) after or without correction of the volume of the sound A signal (see Fig. 6A or Fig. 6B) and the analysis result (second analysis result) after or without correction of the volume of the sound B signal (see Fig. 6A or Fig. 6B). The volume comparison correction unit 16 normalizes the converted values ​​of the volumes of the sound A signal and the sound B signal in step St6-1 to values ​​within a predetermined range (for example, a value between 0 and 1.0) (step St6-2). The volume comparison correction unit 16 smoothes the output values ​​of the volumes of the sound A signal and the sound B signal normalized in step St6-2 (step St6-3).

[0055] The volume comparison correction unit 16 generates a display data signal (see FIG. 3A to FIG. 3C) for visually and comparatively displaying the output values ​​of the volumes of the sound A signal and the sound B signal smoothed in step St6-3 by linking the minimum value of the display bar (see FIG. 3A to FIG. 3C) to 0 (zero) and the maximum value to X (X: a constant where 0≦X≦1) (step St6-4). For example, when X=0.5, the volume comparison correction unit 16 generates a display data signal (the maximum number of display bars) for which the maximum value is a value where the volume output value is 0.5 or more. On the other hand, when X=0.5, for example, the volume comparison correction unit 16 generates the number of display bars where the volume output value is 0.4. The user SP1 only needs to know whether the volume of the voice uttered by him / herself among the input sound signals is sufficient to be transmitted to the other party of the remote call, not whether the voice uttered by him / herself is the maximum. For this reason, the volume comparison and correction unit 16 can make the most of the resolution of the display bar by introducing X (see above) which serves as a reference value.

[0056] Next, a detailed example of the operation procedure of the parameter calculation process in step St7 in Fig. 4 will be described with reference to Fig. 8. Each process shown in Fig. 8 is mainly executed by the control unit 17 of the processor PRC1 (see Fig. 2).

[0057] In FIG. 8, the control unit 17 determines whether or not a specific second condition ("B≧bY?") is satisfied, that is, whether or not the volume of the sound B signal analyzed in step St4 of FIG. 4 is equal to or greater than a certain value (an example of a second threshold value) preregistered in the memory 11 (step St7-1). This certain value indicates, for example, a noise volume (level) that interferes with the hearing of the speech of the user SP1, who is the speaker, by the call partner (i.e., users SP2, SP3, SP4) during a remote call. When the control unit 17 determines that the volume of the sound B signal is equal to or greater than the certain value (an example of a second threshold value) (step St7-1, YES), it increments the counter c (i.e., adds a predetermined value i to the counter c) (step St7-2). As a result, the counter c←c+i. The predetermined value i is a constant value, for example, 2.

[0058] On the other hand, when the control unit 17 determines that the volume of the sound B signal is less than a certain value (an example of a second threshold value) (step St7-1, NO), it decrements the counter c (i.e., subtracts a predetermined value i from the counter c) (step St7-3). As a result, the counter c←ci.

[0059] After step St7-2 or step St7-3, the control unit 17 judges whether the value of the counter c at that time is equal to or greater than the certain value Z preregistered in the memory 11 (step St7-4). If it is judged that the value of the counter c is less than the certain value Z (step St7-4, NO), the process of the control unit 17 in Fig. 8 ends. In other words, if the volume of the sound B signal does not remain equal to or greater than the certain value bY for a certain period of time, the control unit 17 ends the process of Fig. 8 without performing the message display process of step St7-5.

[0060] On the other hand, when the control unit 17 determines that the value of the counter c is equal to or greater than the certain value Z (step St7-4, YES), it executes a message display process to display a predetermined message (for example, "There is a problem due to noise generation") on the display DP1 (step St7-5). As a result, since the volume of the sound B signal has been equal to or greater than the certain value bY for a certain period of time, the control unit 17 executes the message display process of step St7-5 to alert the user SP1, who is the speaker of the remote call, to the situation of surrounding noise, or to urge the user to improve the situation.

[0061] As described above, in the telephone call system 100 according to the first embodiment, the sound processing device 10 separates an input sound signal collected by a sound collection device (for example, a headset HS1) into a noise sound signal and a non-noise sound signal, analyzes the volume of the non-noise sound signal, and analyzes the volume of the noise sound signal. The sound processing device 10 visually and contrastively displays a first analysis result, which is an analysis result of the volume of the non-noise sound signal, and a second analysis result, which is an analysis result of the volume of the noise sound signal, on the display DP1. This allows the sound processing device 10 to explicitly notify the extent to which noise sound is present around the user SP1, which is a speaker of a remote call, and which is mixed in during execution of a specific process (for example, during a call). Therefore, the user SP1 can easily and quantitatively grasp whether or not the surrounding noise sound is hindering the continuation of the remote call in the environment in which the user SP1 is speaking, and can improve the situation (for example, reduce or move the noise source) or inform the call partner of the surrounding situation.

[0062] <Embodiment 2> In the second embodiment, as one implementation of the noise analysis device according to the present disclosure, a voiceprint authentication system that uses the speaker's speech to authenticate the speaker in order to determine whether the speaker should open or close a door will be exemplified (see FIG. 9). Note that in the description of the second embodiment, the same configurations and operations as those of the first embodiment will be given the same reference numerals, and the description will be simplified or omitted, and only differences will be described.

[0063] First, a system configuration example of a voiceprint authentication system 100A according to the second embodiment will be described with reference to Fig. 9. Fig. 9 is a diagram showing a system configuration example of the voiceprint authentication system 100A according to the second embodiment. The voiceprint authentication system 100A includes at least a sound processing device 10A connected to each other via a network NW1 so as to be capable of mutual data communication. A display DP1 for display is connected to the sound processing device 10A. The sound processing device 10A is provided with a sound collection device MC1 (see Fig. 10) that collects the speech of a user SP1 who is a speaker.

[0064] The door opening / closing control device DRC1 is connected to the network NW1 and controls the opening / closing of at least one door DR1 installed at a predetermined position of a store, facility, etc., based on the processing result from the voiceprint authentication server 50. Specifically, when the door opening / closing control device DRC1 receives information from the voiceprint authentication server 50 indicating that the voiceprint authentication was successful, it controls the door DR1 to open. On the other hand, when the door opening / closing control device DRC1 receives information from the voiceprint authentication server 50 indicating that the voiceprint authentication was unsuccessful, it controls the door DR1 to remain closed without opening it.

[0065] The voiceprint authentication server 50 is connected to the network NW1, and when it receives the speech signal of the user SP1 sent from the sound processing device 10A, it performs voiceprint authentication to determine whether the voice corresponding to the speech signal is the voice of the user SP1 by using the speech signal and a predetermined voiceprint authentication database (not shown). Only when it is determined that the voice feature amount included in the speech signal sent from the sound processing device 10A is the same as the voice feature amount of the user SP1 preregistered in the voiceprint authentication database, the voiceprint authentication server 50 returns information indicating that the voiceprint authentication has been successful to at least one of the door opening / closing control device DRC1 and the sound processing device 10A. In addition, when the voiceprint authentication server 50 determines, as part of voiceprint authentication, that the voice features contained in the speech voice signal sent from the sound processing device 10A are dissimilar to the voice features of the user SP1 pre-registered in the voiceprint authentication database, the voiceprint authentication server 50 returns information to the effect that the voiceprint authentication has failed to at least one of the door opening / closing control device DRC1 and the sound processing device 10A.

[0066] Next, with reference to FIG. 10, a hardware configuration example of the sound processing device 10A of FIG. 9 will be described. FIG. 10 is a block diagram showing a hardware configuration example of the sound processing device of FIG. 9. The sound processing device 10A is a general-purpose computer device, and may be, for example, a personal computer, a smartphone, a tablet terminal, or a server computer. The sound processing device 10A includes at least a sound collection device MC1, a processor PRC1A, and a memory 11 as a hardware configuration. The sound processing device 10A may further include an input device such as a mouse that accepts an operation of a user SP1. The display DP1 may be included as a part of the sound processing device 10A.

[0067] The sound collection device MC1 is, for example, a well-known microphone, which collects the speech of the user SP1 and sends it to the processor PRC1A. The signal of this speech (input sound signal) is input to each of the sound analysis unit 13 and the voiceprint authentication unit 18A.

[0068] Here, a detailed description will be given of an example of the configuration of the executing entity of the processing functionally executed by the processor PRC1A. The processor PRC1A functionally includes a sound analysis unit 13, a sound A volume analysis unit 14, a sound B volume analysis unit 15, a volume comparison and correction unit 16, a control unit 17A, a voiceprint authentication unit 18A, and an authentication control unit 19A.

[0069] The control unit 17A judges whether or not the second analysis result input from the volume comparison and correction unit 16 satisfies a specific third condition (see FIG. 8). When the control unit 17 judges that the second analysis result satisfies the specific third condition, the control unit 17 generates information to the effect that the sound A signal corresponding to the speech of the user SP1 will not be used for voiceprint authentication, or performs processing to convert the volume (level) of the sound A signal to zero. The control unit 17A sends information to the voiceprint authentication unit 18A to the effect that the sound A signal corresponding to the speech of the user SP1 will not be used for voiceprint authentication, or sends the sound A signal after processing to convert the volume (level) of the sound A signal to zero to the voiceprint authentication unit 18A. Details of the processing of the control unit 17A will be described later with reference to FIG. 11.

[0070] When the voiceprint authentication unit 18A receives from the control unit 17A information that the sound A signal will not be used for voiceprint authentication or the sound A signal after conversion to a volume (level) of zero, it determines not to perform voiceprint authentication using the sound A signal. On the other hand, when the voiceprint authentication unit 18A does not receive from the control unit 17A information that the sound A signal will not be used for voiceprint authentication or the sound A signal after conversion to a volume (level) of zero, it sends the input sound signal from the sound collection device MC1 to the authentication control unit 19A as a signal for voiceprint authentication.

[0071] The authentication control unit 19A sends the signal from the voiceprint authentication unit 18A to the voiceprint authentication server 50 via the network NW1 as a spoken voice signal for voiceprint authentication of the user SP1. When the authentication control unit 19A receives information from the voiceprint authentication server 50 that the voiceprint authentication has been successful, it displays that fact on the display DP1. When the authentication control unit 19A receives information from the voiceprint authentication server 50 that the voiceprint authentication has been successful, it may send a message to the door opening / closing control device DRC1 to open the door DR1. On the other hand, when the authentication control unit 19A receives information from the voiceprint authentication server 50 that the voiceprint authentication has failed, it displays that fact on the display DP1.

[0072] In the second embodiment, the sound processing device 10A executes the processes of Fig. 4, Fig. 5, Fig. 6A, Fig. 6B, and Fig. 7 corresponding to the operation procedures of the sound processing device 10 according to the first embodiment, and also executes the process of Fig. 11. Here, with reference to Fig. 11, another detailed example of the operation procedures of the parameter calculation process of step St7 in Fig. 4 will be described. Fig. 11 is a flowchart showing in chronological order the detailed example of the operation procedures of the parameter calculation for voiceprint authentication in step St7 in Fig. 4 according to the second embodiment. The processes shown in Fig. 11 are mainly executed by the control unit 17A of the processor PRC1 (see Fig. 2).

[0073] In FIG. 11, the control unit 17A judges whether or not a specific third condition ("B≧bZ?") is satisfied, that is, whether or not the volume of the sound B signal analyzed in step St4 in FIG. 4 is equal to or greater than a certain value (an example of a third threshold value) preregistered in the memory 11 (step St7-11). This certain value indicates, for example, the volume (level) of a noise sound that affects the success or failure of authentication when the voice of the speaker, user SP1, executes a voiceprint authentication process. When the control unit 17A judges that the volume of the sound B signal is equal to or greater than the certain value (an example of a third threshold value) (step St7-11, YES), it generates information that voiceprint authentication will not be performed, or converts the volume (level) of the sound A signal for voiceprint authentication to zero (step St7-12). In other words, if the volume of the sound B signal is equal to or greater than the certain value bZ, the control unit 17A determines that the situation is not suitable for voiceprint authentication, and therefore generates a sound A signal that is not suitable for voiceprint authentication or does not execute voiceprint authentication, and ends the process of FIG. 11.

[0074] On the other hand, if it is determined that the volume of the sound B signal is less than a certain value (an example of the second threshold value) (step St7-11, NO), the process of the control unit 17A in Fig. 11 ends. In other words, since this situation is assumed to be an environment suitable for voiceprint authentication, the control unit 17A instructs the voiceprint authentication unit 18A to use the input sound signal from the sound collection device MC1 (in other words, the sound A signal) for voiceprint authentication. This ends the process of Fig. 11.

[0075] As described above, in the voiceprint authentication system 100A according to the second embodiment, the sound processing device 10A separates an input sound signal collected by the sound collection device MC1 into a noise sound signal and a non-noise sound signal, analyzes the volume of the non-noise sound signal, and analyzes the volume of the noise sound signal. The sound processing device 10A visually and contrastively displays a first analysis result, which is an analysis result of the volume of the non-noise sound signal, and a second analysis result, which is an analysis result of the volume of the noise sound signal, on the display DP1. Furthermore, when the volume of the noise sound signal is greater than a certain value, the sound processing device 10A assumes that the environment is not suitable for voiceprint authentication processing, and therefore stops execution of voiceprint authentication. In this way, the sound processing device 10A can explicitly notify the extent to which noise sound exists around the user SP1, who is a subject of voiceprint authentication, that has been mixed in while a specific process is being performed by the user SP1 (for example, during a call for voiceprint authentication). Therefore, user SP1 can easily and quantitatively grasp whether or not ambient noise in the environment in which he or she is speaking is interfering with the execution of voiceprint authentication, and can take measures to improve the situation (e.g., reducing or moving the noise source) or perform voiceprint authentication properly in a quiet environment.

[0076] <First Modification of First Embodiment> In the first modification of the first embodiment, as one embodiment of the noise analysis device according to the present disclosure, a call system that supports remote calls between multiple people using a sound processing device and a headset connected to each other via a network so that data communication can be performed is illustrated (see FIG. 12), and an example in which the sound processing device actively removes noise is described. Note that in the description of the first modification of the first embodiment, the same reference numerals are given to the same configurations and operations as those of the first embodiment, and the description is simplified or omitted, and differences are described.

[0077] The configuration example of the telephone call system according to the first modification of the first embodiment is the same as the telephone call system 100 according to the first embodiment shown in FIG.

[0078] Next, with reference to FIG. 12, a hardware configuration example of a sound processing device 10B of a call system according to a first modification of the first embodiment will be described. FIG. 12 is a block diagram showing a hardware configuration example of the sound processing device 10B according to a first modification of the first embodiment. The sound processing device 10B is a general-purpose computer device, and may be, for example, a personal computer, a smartphone, a tablet terminal, or a server computer. The sound processing device 10B includes at least a processor PRC1B and a memory 11 as a hardware configuration. The sound processing device 10B may further include an input device such as a mouse that accepts an operation of a user SP1.

[0079] Here, a detailed description will be given of an example of the configuration of the executing entity of the processing functionally executed by the processor PRC1B. The processor PRC1B functionally includes a sound signal input IF 12, a noise removal unit 13B, a noise-removed sound volume analysis unit 14B, a removed sound volume analysis unit 15B, a volume comparison and correction unit 16, a control unit 17, a signal processing unit 18, and a transmission unit 19.

[0080] The noise removal unit 13B separates the input sound signal into at least two types of sound signals (specifically, sound A and sound B) by performing a process of removing noise sounds included in the input sound signal from the sound signal input IF 12, sends the noise-removed sound signal (an example of a non-noise sound signal) to the noise-removed sound volume analysis unit 14B, and sends the noise sound signal (an example of a noise sound signal) to the removed sound volume analysis unit 15B. In the first modification of the first embodiment, sound A indicates the sound of the part of the voice uttered by the user SP1 among the sounds included in the input sound signal, and indicates the input sound signal from which the noise sound has been removed by the noise removal unit 13B. In other words, sound A indicates the sound of the call voice that is assumed to be delivered to the users SP2, SP3, and SP4 who are the call partners by the voice uttered by the user SP1 who is the speaker of the remote call. On the other hand, sound B indicates noise sounds other than the voice uttered by the user SP1 among the sounds included in the input sound signal, and indicates the noise sounds removed by the noise removal unit 13B. In other words, sound B is an unnecessary noise sound that is mixed with sound A delivered to users SP2, SP3, and SP4, and may interfere with users SP2, SP3, and SP4 hearing the speech of user SP1.

[0081] The noise-removed sound volume analysis unit 14B is an example of a first analysis unit, and inputs the noise-removed sound signal from the noise removal unit 13B and analyzes the volume of the input noise-removed sound signal. The noise-removed sound volume analysis unit 14B analyzes the volume of the noise-removed sound signal by referring to, for example, the average power, maximum amplitude, or power in a specific frequency band of the waveform of the noise-removed sound signal. The noise-removed sound volume analysis unit 14B may also smooth the waveform of the input sound A signal in the time direction, and then analyze the volume of the sound A signal by referring to the average power, maximum amplitude, or power in a specific frequency band of the waveform of the sound A signal. The noise-removed sound volume analysis unit 14B sends the analysis result (first analysis result) of the volume of the sound A signal to the volume comparison and correction unit 16.

[0082] The removed sound volume analysis unit 15B is an example of a second analysis unit, and receives the sound B signal from the noise removal unit 13B and analyzes the volume of the received sound B signal. The removed sound volume analysis unit 15B analyzes the volume of the sound B signal by referring to, for example, the average power, maximum amplitude, or power in a specific frequency band of the waveform of the sound B signal. The removed sound volume analysis unit 15B may also smooth the waveform of the input sound B signal in the time direction, and then analyze the volume of the sound B signal by referring to the average power, maximum amplitude, or power in a specific frequency band of the waveform of the sound B signal. The removed sound volume analysis unit 15B sends the analysis result of the volume of the sound B signal (second analysis result) to the volume comparison and correction unit 16.

[0083] Next, an example of an operation procedure of the sound processing device 10B according to the first modification of the first embodiment will be described with reference to Fig. 13 and Fig. 6A to Fig. 8. Fig. 13 is a flowchart showing an example of an operation procedure of the sound processing device 10B according to the first modification of the first embodiment in chronological order. Each process shown in Fig. 6A to Fig. 8 has been described in the first embodiment, so the description will be omitted here.

[0084] An example of the overall operation procedure of the sound processing device 10B according to the first modification of the first embodiment will be described with reference to Fig. 13. Each process shown in Fig. 13 is mainly executed by the processor PRC1 (see Fig. 2).

[0085] 13, the processor PRC1B holds the input sound signal input from the headset HS1 for a certain period of time in the sound signal input IF 12 (step St1). The certain period of time is, for example, 1 second, 5 seconds, 10 seconds, etc. The processor PRC1B uses the input sound signal for the certain period of time held in step St1 to perform a noise sound removal process for removing noise contained in the input sound signal, and separates the input sound signal into a sound A signal and a sound B signal in the noise removal unit 13B and outputs them (step St11).

[0086] The processor PRC1B uses the sound A signal obtained in step St11 to analyze the volume of the sound A signal in the noise-removed sound volume analysis unit 14B (step St3B). The processor PRC1B uses the sound B signal obtained in step St11 to analyze the volume of the sound B signal in the noise-removed sound volume analysis unit 15B (step St4B). The processor PRC1B inputs the analysis result of the volume of the sound A signal obtained by the analysis in step St3B (first analysis result) and the analysis result of the volume of the sound B signal obtained by the analysis in step St4B (second analysis result) into the volume comparison and correction unit 16 for comparison (step St5). A detailed example of the operation procedure of the volume comparison and correction process in step St5 has been described with reference to FIGS. 6A and 6B, and therefore will not be described here.

[0087] The processor PRC1B performs a volume display process (see FIG. 7) using the analysis result (first analysis result) of the volume of the sound A signal obtained by the analysis in step St3B or the output after correction in step St5, and the analysis result (second analysis result) of the volume of the sound B signal obtained by the analysis in step St4B or the output after correction in step St5, thereby generating a display data signal for visually and comparatively displaying the respective volumes of the sound A signal and the sound B signal (step St6). A detailed example of the operation procedure of the volume display process in step St6 has been described with reference to FIG. 7, so a description thereof will be omitted here.

[0088] After the volume display process in step St6, the processor PRC1B performs a signal processing parameter calculation process (step St7) using the analysis result (second analysis result) of the volume of the sound B signal obtained by the analysis in step St4B. A detailed operation procedure example of the calculation process in step St7 has been described with reference to FIG. 8, so a description thereof will be omitted here.

[0089] After step St7, if the end of the remote call is instructed by the input operation of the user SP1 (step St8, YES), the process of the processor PRC1B shown in Fig. 13 ends. On the other hand, if the end of the remote call is not instructed by the input operation of the user SP1 (step St8, NO), the process of the processor PRC1B returns to step St1. In other words, the processor PRC1B repeatedly executes a series of processes from step St1 to step St8 until the end of the remote call is instructed by the input operation of the user SP1.

[0090] As described above, in the communication system 100 according to the first modification of the first embodiment, the sound processing device 10B separates an input sound signal collected by a sound collection device (for example, a headset HS1) into a noise sound signal and a non-noise sound signal by a noise sound removal process, analyzes the volume of the non-noise sound signal, and analyzes the volume of the noise sound signal. The sound processing device 10B visually and contrastively displays a first analysis result, which is an analysis result of the volume of the non-noise sound signal, and a second analysis result, which is an analysis result of the volume of the noise sound signal, on the display DP1. This allows the sound processing device 10B to explicitly notify the extent to which noise sound is present around the user SP1, which is a speaker of a remote call, and which is mixed in during execution of a specific process (for example, during a call). Therefore, the user SP1 can easily and quantitatively grasp whether or not the surrounding noise sound is hindering the continuation of the remote call in the environment in which the user SP1 is speaking, and can improve the situation (for example, reduce or move the noise source) or inform the call partner of the surrounding situation.

[0091] <Modification 2 of the First Embodiment> In the second modification of the first embodiment, as one embodiment of the noise analysis device according to the present disclosure, a call system that supports remote calls between multiple people using a sound processing device and a headset connected to each other via a network for data communication is illustrated (see FIG. 14), and an example is described in which the sound processing device distinguishably separates sound intervals and silent intervals in an input sound signal. Note that in the description of the second modification of the first embodiment, the same reference numerals are used to designate the same configurations and operations as those of the first embodiment, and the description is simplified or omitted, and differences are described.

[0092] The configuration example of the telephone system according to the second modification of the first embodiment is the same as the telephone system 100 according to the first embodiment shown in FIG.

[0093] Next, with reference to FIG. 14, a hardware configuration example of a sound processing device 10C of a call system according to a second modification of the first embodiment will be described. FIG. 14 is a block diagram showing a hardware configuration example of the sound processing device 10C according to a second modification of the first embodiment. The sound processing device 10C is a general-purpose computer device, and may be, for example, a personal computer, a smartphone, a tablet terminal, or a server computer. The sound processing device 10C includes at least a processor PRC1C and a memory 11 as a hardware configuration. The sound processing device 10C may further include an input device such as a mouse that accepts an operation of a user SP1.

[0094] Here, a detailed description will be given of an example of the configuration of the executing entity of the processing functionally executed by the processor PRC1C. The processor PRC1C functionally includes a sound signal input IF 12, a VAD processing unit Va, a silent section volume analysis unit Vb, a sound analysis unit 13, a sound A volume analysis unit 14, a sound B volume analysis unit 15, a volume comparison correction unit 16, a control unit 17, a signal processing unit 18, and a transmission unit 19.

[0095] The VAD processing unit Va separates the input sound signal from the sound signal input IF 12 into a sound interval and a silent interval in a distinguishable manner. The sound interval here is an interval in which the speech sound of the speaker, user SP1, appears. The silent interval is an interval in which stationary noise such as white noise that occurs steadily appears or is appearing, unlike a sudden noise sound with a high volume (level). The VAD processing unit Va sends the signal of the sound interval included in the input sound signal to the sound analysis unit 13 as a signal including a sound A signal and a sound B signal, and sends the signal of the silent interval included in the input sound signal to the silent interval volume analysis unit Vb.

[0096] The silent section volume analysis unit Vb is an example of a third analysis unit, and receives an input sound signal (silent section signal) of a silent section from the VAD processing unit Va, and analyzes the volume of the input silent section signal. The silent section volume analysis unit Vb analyzes the volume of the silent section signal, for example, by referring to the average power, maximum amplitude, or power in a specific frequency band of the waveform of the silent section signal. The silent section volume analysis unit Vb may also smooth the waveform of the input silent section signal in the time direction, and then analyze the volume of the silent section signal by referring to the average power, maximum amplitude, or power in a specific frequency band of the waveform of the silent section signal. The silent section volume analysis unit Vb sends the analysis result of the volume of the silent section signal (third analysis result) to the volume comparison and correction unit 16.

[0097] Next, an example of an operation procedure of the sound processing device 10C according to Modification 2 of the embodiment 1 will be described with reference to Fig. 15 and Fig. 6A to Fig. 8. Fig. 15 is a flowchart showing an example of an operation procedure of the sound processing device 10C according to Modification 2 of the embodiment 1 in chronological order. Each process shown in Fig. 6A to Fig. 8 has been described in the embodiment 1, so the description will be omitted here.

[0098] An example of the overall operation procedure of the sound processing device 10C according to the second modification of the first embodiment will be described with reference to Fig. 15. Each process shown in Fig. 15 is mainly executed by the processor PRC1 (see Fig. 2).

[0099] In Fig. 15, the processor PRC1C holds the input sound signal input from the headset HS1 in the sound signal input IF12 for a certain period (step St1). The certain period is, for example, 1 second, 5 seconds, 10 seconds, etc. The processor PRC1C separates the input sound signal into a sound section and a silent section using the input sound signal for the certain period held in step St1 (step St21). If the processor PRC1C determines that a silent section exists in the input sound signal (step St21, NO), it analyzes the volume of the silent section signal, which is the input sound signal of the silent section (step St22). The analysis result of the volume of the silent section signal is referred to and used in step St5.

[0100] On the other hand, when the processor PRC1C determines that a sound interval exists in the input sound signal (step St21, YES), the processor PRC1C performs sound analysis on the sound interval signal using the sound interval signal, which is the input sound signal of the sound interval, and separates the sound interval signal into a sound A signal and a sound B signal in the sound analysis unit 13 and outputs them (step St2). A detailed example of the operation procedure of the sound analysis in step St2 has been described with reference to FIG. 5, so it will not be repeated here.

[0101] The processor PRC1C uses the sound A signal obtained in step St2 to analyze the volume of the sound A signal in the sound A volume analysis unit 14 (step St3). The processor PRC1C uses the sound B signal obtained in step St2 to analyze the volume of the sound B signal in the sound B volume analysis unit 15 (step St4). The processor PRC1C inputs the analysis result of the volume of the sound A signal obtained by the analysis in step St3 (first analysis result) and the analysis result of the volume of the sound B signal obtained by the analysis in step St4 (second analysis result) in the volume comparison correction unit 16 and compares them (step St5). A detailed operation procedure example of the volume comparison correction process in step St5 has been described with reference to each of Figures 6A and 6B, so a description thereof will be omitted here. Note that the volume comparison correction process may be performed by taking into account (for example, adding) not only the second analysis result obtained in step St4 but also the volume analysis result of the silent section signal obtained in step St22 (third analysis result).

[0102] The processor PRC1C performs a volume display process (see FIG. 7) using the analysis result (first analysis result) of the volume of the sound A signal obtained by the analysis in step St3 or the output after correction in step St5, and the analysis result (second analysis result) of the volume of the sound B signal obtained by the analysis in step St4 or the output after correction in step St5, thereby generating a display data signal for visually and comparatively displaying the respective volumes of the sound A signal and the sound B signal (step St6). A detailed example of the operation procedure of the volume display process in step St6 has been described with reference to FIG. 7, so a description thereof will be omitted here.

[0103] After the volume display process in step St6, the processor PRC1C performs a signal processing parameter calculation process (step St7) using the analysis result (second analysis result) of the volume of the sound B signal obtained by the analysis in step St4. A detailed operation procedure example of the calculation process in step St7 has been described with reference to FIG. 8, so a description thereof will be omitted here.

[0104] After step St7, if the end of the remote call is instructed by the input operation of the user SP1 (step St8, YES), the process of the processor PRC1C shown in Fig. 15 ends. On the other hand, if the end of the remote call is not instructed by the input operation of the user SP1 (step St8, NO), the process of the processor PRC1C returns to step St1. In other words, the processor PRC1C repeatedly executes a series of processes from step St1 to step St8 until the end of the remote call is instructed by the input operation of the user SP1.

[0105] As described above, in the communication system 100 according to the second modification of the first embodiment, the sound processing device 10C first separates an input sound signal collected by a sound collection device (for example, a headset HS1) into a sound section signal and a silent section signal by VAD processing, then separates the sound section signal into a noise sound signal and a non-noise sound signal, analyzes the volume of the non-noise sound signal, analyzes the volume of the noise sound signal, and analyzes the volume of the silent section signal. The sound processing device 10C visually and contrastively displays on the display DP1 a first analysis result which is an analysis result of the volume of the non-noise sound signal, a second analysis result which is an analysis result of the volume of the noise sound signal, and a third analysis result which is an analysis result of the volume of the silent section signal. As a result, the sound processing device 10C can explicitly notify the extent to which noise sound around the user SP1, which is a stationary noise or an unexpected noise sound, is present and which has been mixed in during the execution of a specific process (for example, during a call) by the user SP1 who is the speaker of the remote call. Therefore, user SP1 can easily and quantitatively grasp whether or not ambient noise in the environment in which he or she is speaking is interfering with the continuation of the remote call, and can take measures to improve the situation (e.g., reducing or moving the noise source) or inform the other party of his or her surroundings.

[0106] <About the technology disclosed herein> As described above, the present disclosure discloses the following technical ideas.

[0107] <Technology 1> a sound analysis unit (sound analysis unit 13) that separates an input sound signal collected by a sound collection device (headset HS1, sound collection device MC1) into a noise sound signal and a non-noise sound signal; a first analysis unit (sound A volume analysis unit 14) that analyzes the volume of the non-noise sound signal; A second analysis unit (sound B volume analysis unit 15) that analyzes the volume of the noise sound signal; a display control unit (volume comparison and correction unit 16) that visually and comparatively displays a first analysis result, which is an analysis result of the volume of the non-noise sound signal, and a second analysis result, which is an analysis result of the volume of the noise sound signal, on a display (display DP1). Noise analysis equipment. This allows the noise analysis device to explicitly notify the user of the level of noise present around the user that has been mixed in while the user, who is the speaker of a remote call, is performing a specific process (eg, during a call).

[0108] <Technology 2> The input sound signal includes a call voice signal when a user wears the sound collection device and speaks, and a noise sound signal generated when the user speaks. The noise analysis device according to technique 1. This allows the noise analysis device to visually and comparatively display the volume of the call volume signal when the user speaks while wearing a sound collection device such as a headset HS1, and the volume of the noise sound signal occurring around it, in a scene such as a user making a remote call.

[0109] <Technology 3> A comparison unit (volume comparison and correction unit 16) that compares the first analysis result with the second analysis result, the comparison unit corrects the volume of the non-noise sound signal to be smaller than the volume of the first analysis result when the volume of the noise sound signal indicated by the second analysis result is equal to or larger than the product of the volume of the non-noise sound signal indicated by the first analysis result and a predetermined coefficient value (e.g., k, k=1, k=0.5). The noise analysis device according to the first or second aspect of the present invention. This enables the noise analysis device to accurately allow the user to understand that the surrounding noise is equal to or greater than the target sound, which is the user's speech, and therefore the target sound is difficult for the other party in the remote call to hear.

[0110] <Technology 4> A comparison unit (volume comparison and correction unit 16) that compares the first analysis result with the second analysis result, the comparison unit corrects the volume of the non-noise sound signal to be smaller than the volume of the first analysis result when the volume of the non-noise sound signal indicated by the first analysis result is equal to or larger than a first threshold value and the volume of the noise sound signal indicated by the second analysis result analyzed immediately before is equal to or larger than the first threshold value. The noise analysis device according to any one of the first to third aspects. This makes it possible for the noise analysis device to prevent the user from being notified of momentary erroneous output of sounds A and B, which may occur frequently.

[0111] <Technology 5> The display control unit normalizes each of the first analysis result and the second analysis result to a value within a predetermined range. The noise analysis device according to any one of the first to fourth aspects. This allows the noise analysis device to display the sound A signal and the sound B signal within a predetermined range of values, making it possible to realize a continuous display.

[0112] <Technology 6> the display control unit displays, as a result of the normalization process, zero as a minimum value and X (X: 0≦X≦1) or more as a maximum value. The noise analysis device according to technology 5. This enables the noise analysis device to inform the user whether the volume of their speech is sufficient for the other party in a remote call to hear, and by introducing the guideline value X, the resolution of the display bar can be utilized to the maximum extent.

[0113] <Technology 7> a control unit (control unit 17) that determines whether a period during which the volume of the noise sound signal indicated by the second analysis result is equal to or greater than a second threshold value is equal to or greater than a certain period, When the control unit determines that the period during which the volume of the noise signal is equal to or greater than the second threshold value is equal to or greater than the certain period, the control unit displays a message on the display notifying the user of the occurrence of a noise sound. The noise analysis device according to any one of the first to sixth aspects. This enables the noise analysis device to alert the user, who is the speaker of the remote call, to a situation in which noise is occurring in the surroundings, and to encourage the user to improve the situation.

[0114] <Technology 8> a determination unit (VAD processing unit Va) that distinguishes between a sound section in the input sound signal and a stationary noise section (a silent section) in the input sound signal; A third analysis unit (silent section volume analysis unit Vb) that analyzes the volume of the stationary noise signal in the stationary noise section, the display control unit visually and contrastively displays on the display a first analysis result which is an analysis result of the volume of the non-noise sound signal, and an addition result of a second analysis result which is an analysis result of the volume of the noise sound signal and a third analysis result which is an analysis result of the volume of the stationary noise signal. The noise analysis device according to any one of the first to seventh aspects. This allows the noise analysis device to explicitly notify the user of the extent to which noise, either steady noise or sudden noise, is present around the user while the user, who is the speaker of a remote call, is performing a specific process (e.g., during a call).

[0115] <Technology 9> The input sound signal includes a speech signal when a user speaks to the sound collection device for voiceprint authentication, and a noise sound signal generated during the speech. The noise analysis device according to any one of the first to eighth aspects. This allows the noise analysis device to explicitly notify the user SP1, who is the subject of voiceprint authentication, of the amount of noise present around the user SP1 that has been mixed in while the user SP1 is performing a specific process (e.g., during a call for voiceprint authentication).

[0116] <Technology 10> a control unit (control unit 17A) that determines whether the volume of the noise sound signal indicated by the second analysis result is equal to or greater than a third threshold value, When the control unit determines that the volume of the noise sound signal is equal to or greater than the third threshold, the control unit interrupts the voiceprint authentication process or converts the level of the speech sound signal for the voiceprint authentication to zero. The noise analysis device according to technology 9. This allows the noise analysis device to determine that the current situation is not suitable for voiceprint authentication, and therefore makes it possible to prevent voiceprint authentication from being performed unnecessarily.

[0117] <Technology 11> 1. A noise analysis method executed by a noise analysis device, comprising: A step of separating an input sound signal collected by a sound collection device into a noise sound signal and a non-noise sound signal; analyzing the loudness of the non-noise sound signal; analyzing the volume of the noise sound signal; and visually and comparatively displaying on a display a first analysis result which is an analysis result of the volume of the non-noise sound signal and a second analysis result which is an analysis result of the volume of the noise sound signal. Noise analysis methods. This allows the noise analysis device to explicitly notify the user of the level of noise present around the user that has been mixed in while the user, who is the speaker of a remote call, is performing a specific process (eg, during a call).

[0118] <Technology 12> The noise analysis device is a computer. A step of separating an input sound signal collected by a sound collection device into a noise sound signal and a non-noise sound signal; analyzing the loudness of the non-noise sound signal; analyzing the volume of the noise sound signal; a step of visually and comparatively displaying on a display a first analysis result which is an analysis result of the volume of the non-noise sound signal and a second analysis result which is an analysis result of the volume of the noise sound signal, program. This allows the noise analysis device to explicitly notify the user of the level of noise present around the user that has been mixed in while the user, who is the speaker of a remote call, is performing a specific process (eg, during a call).

[0119] Although various embodiments have been described above with reference to the accompanying drawings, the present disclosure is not limited to such examples. It is clear that a person skilled in the art can conceive of various modifications, corrections, substitutions, additions, deletions, and equivalents within the scope of the claims, and it is understood that these also belong to the technical scope of the present disclosure. In addition, the components in the various embodiments described above may be arbitrarily combined within the scope of the invention. [Industrial Applicability]

[0120] INDUSTRIAL APPLICABILITY The present disclosure is useful as a noise analysis device, a noise analysis method, and a program that explicitly notify the level of noise present when a user is executing a specific process. [Explanation of symbols]

[0121] 10 Sound Processing Device 10A Sound Processing Device 11. Memory 12 Sound signal input IF 13 Sound Analysis Department 13B Noise Cancellation Section 14 Sound A volume analysis section 14B Noise-removed sound volume analysis section 15 Sound B volume analysis section 15B Removal sound volume analysis section 16 Volume comparison and correction section 17, 17A Control section 18 Signal Processing Section 19 Transmitter 50 Voiceprint authentication server 100 Call System 100A Voiceprint Authentication System DP1 Display MC1 Sound Recording Device NW1 Network PRC1, PRC1A, PRC1B, PRC1C processors SP1, SP2, SP3, SP4 Users Va VAD Processing Unit Vb Silent section volume analysis part

Claims

1. a sound analysis unit that separates an input sound signal collected by a sound collection device into a noise sound signal and a non-noise sound signal; a first analysis unit that analyzes the volume of the non-noise sound signal; a second analysis unit that analyzes the volume of the noise signal; a display control unit that visually and contrastively displays a first analysis result, which is an analysis result of the volume of the non-noise sound signal, and a second analysis result, which is an analysis result of the volume of the noise sound signal, on a display. Noise analysis equipment.

2. The input sound signal includes a call voice signal when a user wears the sound collection device and speaks, and a noise sound signal generated when the user speaks.

2. The noise analysis device according to claim 1.

3. A comparison unit that compares the first analysis result with the second analysis result, the comparison unit corrects the volume of the non-noise sound signal to be smaller than the volume of the first analysis result when the volume of the noise sound signal indicated by the second analysis result is equal to or larger than the product of the volume of the non-noise sound signal indicated by the first analysis result and a predetermined coefficient value.

2. The noise analysis device according to claim 1.

4. A comparison unit that compares the first analysis result with the second analysis result, the comparison unit corrects the volume of the non-noise sound signal to be smaller than the volume of the first analysis result when the volume of the non-noise sound signal indicated by the first analysis result is equal to or larger than a first threshold value and the volume of the noise sound signal indicated by the second analysis result analyzed immediately before is equal to or larger than the first threshold value.

2. The noise analysis device according to claim 1.

5. The display control unit normalizes each of the first analysis result and the second analysis result to a value within a predetermined range.

2. The noise analysis device according to claim 1.

6. the display control unit displays a result of the normalization process with zero as a minimum value and with X (X: 0≦X≦1) or more as a maximum value. The noise analysis device according to claim 5.

7. a control unit that determines whether a period during which the volume of the noise signal indicated by the second analysis result is equal to or greater than a second threshold is equal to or greater than a certain period, when it is determined that the period during which the volume of the noise signal is equal to or greater than the second threshold value is equal to or greater than the certain period, the control unit displays a message on the display notifying the user of the occurrence of a noise sound.

2. The noise analysis device according to claim 1.

8. a determination unit that distinguishes between a sound segment in the input sound signal and a stationary noise segment in the input sound signal; A third analysis unit that analyzes the volume of a stationary noise signal in the stationary noise section, the display control unit visually and contrastively displays on the display a first analysis result which is an analysis result of the volume of the non-noise sound signal, and an addition result of a second analysis result which is an analysis result of the volume of the noise sound signal and a third analysis result which is an analysis result of the volume of the stationary noise signal.

2. The noise analysis device according to claim 1.

9. The input sound signal includes a speech signal when a user speaks to the sound collection device for voiceprint authentication, and a noise sound signal generated during the speech.

2. The noise analysis device according to claim 1.

10. a control unit that determines whether or not a volume of the noise sound signal indicated by the second analysis result is equal to or greater than a third threshold value, When the control unit determines that the volume of the noise sound signal is equal to or greater than the third threshold, the control unit interrupts the voiceprint authentication process or converts the level of the speech sound signal for the voiceprint authentication to zero. The noise analysis apparatus according to claim 9.

11. 1. A noise analysis method executed by a noise analysis device, comprising: A step of separating an input sound signal collected by a sound collection device into a noise sound signal and a non-noise sound signal; analyzing the loudness of the non-noise sound signal; analyzing the volume of the noise sound signal; and visually and comparatively displaying on a display a first analysis result which is an analysis result of the volume of the non-noise sound signal and a second analysis result which is an analysis result of the volume of the noise sound signal. Noise analysis methods.

12. The noise analysis device is a computer. A step of separating an input sound signal collected by a sound collection device into a noise sound signal and a non-noise sound signal; analyzing the loudness of the non-noise sound signal; analyzing the volume of the noise sound signal; a step of visually and comparatively displaying on a display a first analysis result which is an analysis result of the volume of the non-noise sound signal and a second analysis result which is an analysis result of the volume of the noise sound signal, program.

Images