Recording device, recording program, and recording method

Abstract

To provide a sound recording device that can distinguish and record desired voice calls or modem calls.SOLUTION: A sound recording device of the present invention includes: duplicate packet acquisition means for acquiring a duplicate packet of a communication packet transmitted and received between communication terminals from a network; determination means for determining based on a communication sound contained in the duplicate packet whether the communication sound is for a modem call or a voice call; and storage means for sequentially recording the determined communication sound as recording data in a recording unit when a determination result of the determination means is a desired call.SELECTED DRAWING: Figure 1

Description

[Technical field]

[0001] The present invention relates to a recording device, a recording program, and a recording method, and can be applied to a recording device that monitors a communication network in which modem calls and voice calls coexist, and records communication sounds, for example. [Background technology]

[0002] In recent years, IP telephones, modem devices, and the like that use Voice over Internet Protocol (VoIP) technology have become increasingly popular.

[0003] Furthermore, with the widespread use of these devices, various technologies have come into existence for mirroring (duplicating) IP (e.g., RTP (Real-time Transport Protocol)) packets from a communication network and recording (recording) and managing the communication sounds (voice calls and modem calls) contained in the copied packets (see, for example, Patent Documents 1 to 4).

[0004] When the recorded data is a voice call, it is used, for example, in telephone services, etc., to objectively evaluate the voice quality (for example, evaluation using MOS (Mean Opinion Score) values).

[0005] Furthermore, when the recorded data is a modem call, it is used, for example, to analyze whether the control of a device using modem communication is being performed normally. [Prior art documents] [Patent documents]

[0006] [Patent Document 1] JP 2007-282004 A [Patent Document 2] JP 2015-177411 A [Patent Document 3] JP 2015-198358 A [Patent Document 4] JP 2018-174438 A [Patent Document 5] JP 2001-282266 A [Patent Document 6] JP 2011-70084 A [Patent Document 7] JP 2017-184062 A [Patent Document 8] JP 2020-115096 A Summary of the Invention [Problem to be solved by the invention]

[0007] However, in conventional techniques (including the techniques described in the above Patent Documents 1 to 4), even if voice calls and modem calls are mixed together, all of them are recorded without distinguishing between voice calls and modem calls.

[0008] For example, when recording data of modem calls is required for analyzing modem communication problems, recording unnecessary voice calls also causes the following problems.

[0009] First, the capacity of the recording device (for example, a hard disk drive (HDD)) is consumed in large quantities due to the capacity of unnecessary voice calls. In other words, the cost of preparing a large-capacity HDD increases. Second, the amount of recorded data becomes enormous, and since recorded data of voice calls and modem calls is mixed, the time and effort required for analysis increases. Third, the risk of information leaks increases because recorded data of unnecessary voice calls remains on the recording device.

[0010] On the other hand, even if it is desired to only analyze a voice call, problems similar to the above first to third problems arise.

[0011] Therefore, there is a demand for a recording device, a recording program, and a recording method that can distinguish between a desired voice call and a desired modem call and record the call. [Means for solving the problem]

[0012] The first recording device of the present invention comprises: (1) a duplicate packet acquisition means for acquiring a duplicate packet of a communication packet transmitted between communication terminals via a network; (2) a determination means for determining whether a communication sound included in the duplicate packet is a modem call or a voice call; and (3) a determination result of the determination means. The modem call and a storage means for sequentially recording the determined communication sound in a recording unit as recording data when the communication sound is determined to be a communication sound. (4) the determining means determines whether the communication sound is the modem call by detecting a single frequency signal specific to a modem signal. It is characterized by: The second recording device of the present invention comprises (1) a duplicate packet acquisition means for acquiring a duplicate packet of a communication packet transmitted and received between communication terminals from a network; (2) a determination means for determining whether the communication sound is a modem call or a voice call based on the communication sound contained in the duplicate packet; and (3) a storage means for sequentially recording the determined communication sound as recording data in a recording unit when the determination result of the determination means is the voice call, and (4) the determination means generates histogram data regarding signal levels within a sound interval of the communication sound, arranges a predetermined number of the top sound intervals of the histogram data in descending order, and determines whether the communication sound is the voice call based on whether the total value of the number of times the signal level from the highest signal level to a predetermined top level is equal to or greater than a first threshold value and whether the variance value of the histogram data arranged in descending order of signal levels from the highest signal level to a predetermined top level exceeds a second threshold value.

[0013] No. 3 The recording program of the present invention includes a computer including: (1) a duplicate packet acquisition means for acquiring a duplicate packet of a communication packet transmitted between communication terminals via a network; (2) a determination means for determining whether a communication sound included in the duplicate packet is a modem call or a voice call; and (3) a determination result of the determination means. The modem call When the communication sound is determined to be a communication sound, the recording unit functions as a storage unit for sequentially recording the communication sound as recording data. (4) the determining means determines whether the communication sound is the modem call by detecting a single frequency signal specific to a modem signal. It is characterized by: A fourth recording program of the present invention causes a computer to function as (1) a duplicate packet acquisition means for acquiring a duplicate packet of a communication packet transmitted and received between communication terminals from a network, (2) a determination means for determining whether the communication sound is a modem call or a voice call based on the communication sound contained in the duplicate packet, and (3) a storage means for sequentially recording the determined communication sound as recording data in a recording unit when the determination result of the determination means is the voice call, and (4) the determination means generates histogram data regarding signal levels within a sound interval of the communication sound, arranges a predetermined number of the top sound intervals of the histogram data in descending order, and determines whether the communication sound is the voice call based on whether the total value of the number of times the signal level from the highest signal level to a predetermined top level is equal to or greater than a first threshold value and whether the variance value of the histogram data arranged in descending order of signal levels from the highest signal level to a predetermined top level exceeds a second threshold value.

[0014] No. 5 The present invention is a recording method for use in a recording device, comprising: (1) a duplicate packet acquisition method; The step is (2) a determination means for determining whether a communication sound included in the duplicated packet is a modem call or a voice call based on the communication sound included in the duplicated packet; and (3) a storage means for storing the determination result of the determination means. The modem call if so, sequentially recording the determined communication sounds as recording data in a recording unit; (4) The determining means determines whether the communication sound is the modem call by detecting a single frequency signal specific to a modem signal. It is characterized by: A sixth aspect of the present invention is a recording method for use in a recording device, characterized in that (1) a duplicate packet acquisition means acquires a duplicate packet of a communication packet transmitted and received between communication terminals from a network, (2) a determination means determines whether the communication sound is a modem call or a voice call based on the communication sound contained in the duplicate packet, (3) a storage means, when the determination result of the determination means is the voice call, sequentially records the determined communication sound in a recording unit as recording data, and (4) the determination means generates histogram data regarding signal levels within a sound section of the communication sound, arranges a predetermined number of top sound sections of the histogram data in descending order, and determines whether the communication sound is the voice call based on whether the total value of the number of times the signal level from the highest signal level to a predetermined top level is equal to or greater than a first threshold value and whether the variance value of the histogram data arranged in descending order of signal levels from the highest signal level to a predetermined top level exceeds a second threshold value. Effect of the Invention

[0015] The present invention allows for the differentiation of whether a voice call or a modem call is desired to be recorded. [Brief description of the drawings]

[0016] [Figure 1] 1 is a block diagram showing an example of a connection configuration of devices including a sound recording device according to a first embodiment. [Diagram 2] 4 is a flowchart showing the overall operation of the recording device according to the first embodiment. [Diagram 3] 1 is a configuration diagram showing the configuration of a frequency detection device applied to a modem call / voice call determination unit according to a first embodiment. [Figure 4] 4 is an explanatory diagram showing an example of a signal specific to a modem call determined by a modem call / voice call determination unit according to the first embodiment; FIG. [Diagram 5] FIG. 11 is a block diagram showing an example of a connection configuration of devices including a sound recording device according to a second embodiment. [Figure 6] FIG. 11 is an explanatory diagram illustrating an example of a histogram distribution of a signal level of an audio signal according to the second embodiment. [Figure 7] FIG. 11 is an explanatory diagram illustrating an example of a histogram distribution of signal levels of a modem signal according to the second embodiment. [Figure 8] FIG. 11 is a block diagram showing a functional configuration of a voice detection means according to a second embodiment. [Figure 9] 13 is a flowchart showing processing (mainly processing prior to the call section determination) when a call section determination is performed in an RBT removal filter unit and a call section determination unit according to a second embodiment. [Figure 10] 10 is a flowchart showing detailed processing of call section determination in a call section determination unit according to the second embodiment. [Figure 11] 10 is a flowchart showing detailed processing of a signal level histogram generating section and a determination processing section according to the second embodiment. [Figure 12] FIG. 11 is an explanatory diagram showing an example of histogram data of a voice signal and a modem signal according to the second embodiment. [Figure 13] 11 is an explanatory diagram showing a specific example of silencing a PB signal including a PB signal silencing device according to a modified embodiment. FIG. [Figure 14] FIG. 13 is a configuration diagram showing an example of a PB signal detector according to a modified embodiment. [Figure 15] FIG. 13 is a block diagram showing an example of a dial pulse signal muting device according to a modified embodiment. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0017] (A) First embodiment Hereinafter, a first embodiment of a sound recording device, a sound recording program, and a sound recording method will be described in detail with reference to the drawings.

[0018] (A-1) Configuration of the First Embodiment (A-1-1) Overall composition FIG. 1 is a block diagram showing an example of a connection configuration of devices including a sound recording device according to the first embodiment.

[0019] FIG. 1 shows a configuration in which a communication terminal 10 is connected to a fixed telephone line (analog telephone line), and the fixed telephone line is connected to a communication terminal 60 via an IP network 50 (VoIP network).

[0020] The types of the communication terminals 10 and 60 (for example, telephone sets and modem terminals (including FAX and the like)) and the corresponding communication methods are not limited.

[0021] In the configuration of Fig. 1, a fixed telephone line (analog telephone line) connected to a communication terminal 10 is connected to an IP network 50 via an exchange 20, a gateway device 30, and a switch 40 (SW40). In Fig. 1, the connection configuration between the IP network 50 and the communication terminal 60 is not shown, but various connection configurations can be applied.

[0022] The exchange 20 receives an analog signal supplied from the near-end side (communication terminal 10), converts it to digital form and sends it to the transmission path (the transmission path of the fixed telephone line) on the far-end side. The exchange 20 also converts a digital signal supplied from the transmission path on the far-end side (communication terminal 60) to an analog signal and supplies it to the communication terminal 10.

[0023] The gateway device 30 is a relay device that connects a fixed telephone line to an IP network 50. The gateway device 30 may be provided with an adaptive echo canceller or the like for controlling echoes on the near-end side when terminating the fixed telephone line on the near-end side.

[0024] The switch 40 is a relay device disposed between the gateway device 30 and the IP network 50. The switch 40 also acquires packets (e.g., RTP packets) including communication sounds (voice calls, modem calls) communicated between the communication terminals 10 and 60, and transfers them to the recording device 80 described below. There are no particular limitations on the method of acquiring the packets, but for example, the switch 40 may receive packets output from a mirroring port and extract RTP packets related to the communication terminals 10 and 60 to be monitored.

[0025] The recording device 80 is a device that records either a voice call or a modem call among packets transferred from the switch 40. In this embodiment, the recording device 80 records either a voice call or a modem call among communication packets between the communication terminal 10 and the communication terminal 60 in real time, but it is not necessary to do this in real time, and either a voice call or a modem call may be recorded by providing data (packets) at a predetermined timing.

[0026] (A-1-2) Detailed configuration of recording device 80 In FIG. 1, a recording device 80 has a primary recording section 81, a modem call / voice call determination section 82, a recording data storage section 83, a recording section 84, and a communication section 85 as a duplicate packet acquisition means.

[0027] The recording device 80 may be realized by installing a program (a recording program according to the embodiment) in a computer having a processor, memory, etc., but even in this case, the recording device 80 can be functionally shown in Fig. 1. Note that the recording device 80 may be realized in part or in whole by hardware.

[0028] The primary recording unit 81 records the communication sound contained in the mirrored packet transferred from the switch 40 in a primary recording medium. The primary recording medium is, for example, a memory, a HDD, etc. The data recorded in the primary recording unit 81 is deleted at a predetermined timing, regardless of the judgment result of the modem call / voice call judgment unit 82 described later. Note that, as a modified example, the primary recording unit 81 may be omitted. In this case, the communication sound acquired from the SW 40 is input as an input signal directly to the modem call / voice call judgment unit 82 described later.

[0029] The modem call / voice call determination unit 82 determines whether the communication sound recorded in the primary recording unit 81 is a modem call or a voice call.

[0030] In the first embodiment, the modem call / voice call determination unit 82 determines whether the communication sound is a modem call or a voice call depending on whether a signal specific to a modem call has been detected. The signal specific to a modem call is a response signal (2100 Hz) from the receiving modem or a single-frequency signal that appears in modem communications in the ITU-T Recommendation V series. The modem call / voice call determination unit 82 provides the determination result of the communication sound to the recording data storage unit 83, which will be described later.

[0031] When the result of the determination by the modem call / voice call determination unit 82 is a desired call, the recorded data storage unit 83 records the communication sound (recorded data) in the recording unit 84.

[0032] The recorded data storage unit 83 basically separates two-way audio (audio data between the communication terminal 10 and the communication terminal 60) and records it in the recording unit 84, but it may also add and synthesize the two-way audio data and record it in the recording unit 84. This is because it saves the HDD capacity of the recording unit 84.

[0033] The recording unit 84 stores a communication sound (modem call or voice call) that is determined to be a desired call by the recorded data storage unit 83. The recording unit 84 is configured with, for example, a HDD or the like.

[0034] The communication unit 85 communicates with the switch 40 and acquires the mirrored communication packets (communication sound).

[0035] (A-2) Operation of the First Embodiment Next, the operation of the recording device 80 of the first embodiment having the above configuration will be described.

[0036] (A-2-1) Overall operation of recording device 80 FIG. 2 is a flowchart showing the overall operation of the recording device according to the first embodiment.

[0037] The recording device 80 acquires (receives) the communication sound contained in the packets between the communication terminal 10 and the communication terminal 60 mirrored by the SW 40 via the communication unit 85 (S11).

[0038] The recording device 80 records the acquired communication sound in the primary recording unit 81 (S12).

[0039] The modem call / voice call determination unit 82 of the recording device 80 determines whether the communication sound recorded in the primary recording unit 81 is a modem call or not (S13). The modem call / voice call determination unit 82 provides the determination result to the recording data storage unit 83. In the first embodiment, if the communication sound is determined not to be a modem call, it is regarded as a voice call.

[0040] The recording data storage unit 83 of the recording device 80 judges whether the result of the judgment in the above-mentioned step S13 matches the desired call (S14). Note that the desired call here is judged based on data recorded in advance in a setting file or the like (data indicating whether a modem call or a voice call is to be stored).

[0041] The recording data storage unit 83 of the recording device 80 records the communication sound as recording data in the recording unit 84 only when the determined communication sound matches the desired call (S15).

[0042] (A-2-2) Modem call specific signal detection operation (detailed operation of step S13) Next, a method for detecting a signal specific to a modem call by the modem call / voice call determination unit 82 will be described. Here, a method for detecting a single frequency signal (a signal specific to a modem call) in the time domain will be described (otherwise, for example, the technology described in Patent Document 8 can also be applied).

[0043] FIG. 3 is a configuration diagram showing the configuration of a frequency detection device applied to the modem call / voice call determination unit according to the first embodiment.

[0044] 3, the band pass filter (BPF) 110 is set in advance so that a desired selected band signal can be transmitted at 0 dB. If a selected band signal that can transmit 0 dB through the BPF 110 is "S" and an out-of-selection band signal is "N," the input signal can be expressed as "S+N" and the BPF output signal as "S."

[0045] The determiner 120 calculates the value of the average power POW_S of the selected band signal S and the value of the average power POW_N of the out-of-band signal N, and further calculates the signal-to-noise ratio SNR.

[0046] When the value of the average power POW_S of the selected band signal S is greater than or equal to a predefined threshold TH_POW_S and the signal-to-noise ratio SNR is greater than or equal to a predefined threshold TH_SNR, the input signal can be determined to be a single-frequency signal that can pass through the BPF.

[0047] This can be specifically expressed by the following formula: POW_S= 2 > …(A1) POW_N=<(S+N) 2 >-POW_S = 2 +2SN+N 2 >-POW_S = 2 >+2 <sn>+ <N 2 >-POW_S = 2 >+ <N 2 >-POW_S = <N 2 > …(A2) SNR = 10log 10 (POW_S / POW_N) …(A3)

[0048] If POW_S≧TH_POW_S and SNR≧TH_SNR, the input signal is the selected single frequency. Otherwise, the input signal is not the selected single frequency.

[0049] In the above formula, the operator <> represents the time average, and is the average value of the product SN of the single-frequency signal S and the other signal N. <sn>is zero.

[0050] FIG. 4 is an explanatory diagram showing an example of a signal specific to a modem call that is determined by the modem call / voice call determination unit according to the first embodiment.

[0051] The modem call / voice call determination unit 82, for example, sets a single frequency signal used in a desired modem communication standard among the modem communication standards shown in Fig. 4 in the frequency detection device 100 shown in Fig. 3 described above, and determines the modem signal. When there are multiple modem signals to be determined, the frequency detection devices 100 may be configured in parallel.

[0052] In addition, the 2100 Hz single frequency signal may undergo phase inversion at 450 ms intervals or AM modulation, but these may not be distinguished. Alternatively, a log indicating the signal type may be recorded in the recording unit 84 together with the recording data.

[0053] (A-3) Effects of the First Embodiment According to the first embodiment, the following effects can be obtained.

[0054] The recording device 80 records only the necessary calls (voice calls or modem calls) from among the communication sounds in the recording unit 84 as recording data, thereby making it possible to save recording capacity.

[0055] Furthermore, by recording only the desired call in the recording unit 84, the recorded data of voice calls and modem calls are no longer mixed together, making it possible to smoothly analyze either call.

[0056] Furthermore, since unnecessary data that is not related to the desired call is not recorded in the recording unit 84, the risk of information leakage can be reduced.

[0057] (B) Second embodiment Hereinafter, a second embodiment of a sound recording device, a sound recording program, and a sound recording method will be described in detail with reference to the drawings.

[0058] (B-1) Configuration of the second embodiment (B-1-1) Overall composition Fig. 5 is a block diagram showing an example of a connection configuration of each device including a sound recording device according to the second embodiment. The difference from the connection configuration in Fig. 1 (the connection configuration in the first embodiment) is that in Fig. 5, sound recording device 80A is applied instead of sound recording device 80.

[0059] 5, recording device 80A has a primary recording unit 81, a modem call / voice call determination unit 82A, a recorded data storage unit 83, a recording unit 84, and a communication unit 85. Recording device 80A differs from recording device 80 of the first embodiment in that a modem call / voice call determination unit 82A is applied instead of the above-mentioned modem call / voice call determination unit 82. The following description will focus on the differences from the first embodiment.

[0060] The modem call / voice call determination unit 82A of the second embodiment includes a voice detection device 70. Because a voice signal has a characteristic that the signal level fluctuates more significantly than other signals (such as a modem signal), the voice detection device 70 determines whether an input signal (communication sound) is a voice signal (voice call) (in other words, whether it is a voice signal or a modem signal) based on a histogram distribution of the signal level.

[0061] Fig. 6 is an explanatory diagram illustrating an example of a histogram distribution of the signal level of an audio signal according to the second embodiment. Fig. 6(A) shows an example of the waveform of an audio signal for a certain period of time (8 seconds). Fig. 6(B) shows the signal level of the audio signal in Fig. 6(A) in a 5 ms cycle, with the horizontal axis representing time and the vertical axis representing signal level (signal power). Fig. 6(C) shows a histogram of the number of times for each signal level in Fig. 6(B), with the horizontal axis representing the signal level axis and the vertical axis representing the number of times for each signal level.

[0062] On the other hand, Fig. 7 is an explanatory diagram illustrating an example of a histogram distribution of the signal level of a modem signal according to the second embodiment. Fig. 7(A) to (C) show each piece of information of a modem signal (G3 FAX) in the same manner as Fig. 6(A) to (C) described above.

[0063] If we compare the histogram distribution of the signal level of a voice signal (Figure 6(C)) with the histogram distribution of the signal level of a modem signal (Figure 7(C)), we can see that the distribution of the voice signal in the histogram is scattered, whereas the modem signal is concentrated at a constant signal level.

[0064] Therefore, in the second embodiment, after performing a voice determination for an input signal, a voice section of a predetermined time is extracted to generate a histogram of signal levels, and the input signal is determined to be voice or not by referring to the variance value of the histogram, etc. This makes it possible to detect voice with a smaller amount of calculation than, for example, using the method (autocorrelation) described in Patent Document 5.

[0065] (B-1-2) Detailed configuration of the voice detection unit FIG. 8 is a block diagram showing a functional configuration of the voice detection device according to the second embodiment.

[0066] 8, the voice activity detection device 70 includes an RBT removal filter unit 71, a speech interval determination unit 72, a silent interval removal / speech activity extraction unit 73, a signal level histogram generation unit 74, and a determination processing unit 75.

[0067] The RBT removal filter unit 71 is a filter that removes the frequency band (400±20 Hz) of the ring back tone (RBT) from the input signal. Since dial tones and busy tones also have a frequency band similar to that of the RBT, the corresponding frequency bands are removed by this filter.

[0068] The speech interval determination unit 72 is a functional unit that determines the speech interval for the input signal. For example, in Japan, the speech interval determination unit 72 focuses on the nature of RBT being executed by repeating 1 second on and 2 seconds off, and determines the RBT interval from the difference between the signal level of the input signal provided from the primary recording unit 81 and the signal level of the input signal filtered by the RBT removal filter unit 71, and determines the period immediately after the end of the RBT interval as the speech interval. Furthermore, if no RBT is detected for a predetermined time (e.g., 8 seconds), the speech interval determination unit 72 determines that there is no RBT interval. Then, the speech interval determination unit 72 sends the determination result to the silent interval removal and sound activity extraction unit 73.

[0069] The silent interval removal / sound extraction unit 73 removes silent intervals from the call interval determined by the call interval determination unit 72, and extracts a sound interval of a predetermined time. Since the call interval determined by the call interval determination unit 72 means an interval other than the RBT interval, the silent interval removal / sound extraction unit 73 first removes silent intervals from the call interval where it can be determined that the speaker is not speaking. The silent interval removal / sound extraction unit 73 can apply various methods for removing silent intervals from the call interval, but for example, a method of performing a sound / silence determination using the technology described in Patent Document 6, determining a silent interval within the call interval, and removing the determined silent interval from the call interval, or the like, may be used.

[0070] The silent section removal / sound extraction unit 73 extracts a sound section of a predetermined time (e.g., 8 seconds) from the remaining sound section obtained by removing the silent section from the call section (hereinafter, the extracted section will be referred to as the “extracted sound section”). The silent section removal / sound extraction unit 73 provides information on the extracted extracted sound section (e.g., information on the signal level within the extracted sound section) to the signal level histogram generation unit 74.

[0071] The signal level histogram generating unit 74 accumulates the number of signal levels per unit time (for example, 5 ms) in the extracted sound segment to generate histogram data. The signal level histogram generating unit 74 provides the generated histogram data to the determination processing unit 75.

[0072] The determination processing unit 75 arranges the top n (n is 15, for example) of the histogram data (extracted sound intervals) given by the signal level histogram generating unit 74 in descending order of sound (for example, -40 dBm or more). When there are less than n pieces of histogram data showing sound, less than n pieces of histogram data are used. When the total value (total number of occurrences) calculated using the histograms arranged in descending order is less than a threshold T1 (for example, several hundred), it is determined that there is no sound (or the sound interval is short). Furthermore, when the variance value (div) calculated using the histograms arranged in descending order is greater than a threshold T2 (for example, 35), the determination processing unit 75 determines that it is a voice signal, and when it is less than the threshold T2, it determines that it is a signal other than voice, such as a modem, FAX, DTMF, or hold tone. The determination processing unit 75 gives the determination result to the recording data storage unit 83.

[0073] (B-2) Operation of the Second Embodiment Next, the operation of the recording device 80A of the second embodiment having the above configuration will be described.

[0074] The overall operation of the recording device 80A of the second embodiment is similar to that of the recording device 80 shown in Fig. 2. However, the process of step S13 (processing for determining whether the communication sound is present) is different from that of the first embodiment, and the following description will focus on the differences.

[0075] The modem call / voice call determination unit 82A (voice detection device 70) starts processing when it acquires a communication sound (input signal S1) from the primary recording unit 81. The input signal S1 is given in units divided into frame periods of a predetermined length (e.g., 5 ms). Hereinafter, when the divided input signal S1 is expressed in frame units, it will be called S1(f). In other words, "(f)" indicates the parameter value in the fth frame (the same applies to S2, P1, P2, P1', P2', and L described later).

[0076] FIG. 9 is a flowchart showing processing (mainly processing prior to the call section determination) when the call section determination is performed in the RBT removal filter unit and the call section determination unit according to the second embodiment.

[0077] First, the RBT elimination filter unit 71 removes the RBT frequency component (frequency 400±20 Hz) from the input signal S1(f). Hereinafter, the input signal resulting from the filtering process of the RBT elimination filter unit 71 on the input signal S1(f) will be referred to as the “RBT elimination filter passed signal S2(f).” The RBT elimination filter unit 71 provides the RBT elimination filter passed signal S2(f) to the call section determination unit 72.

[0078] The speech section determination unit 72 first calculates the signal level for each of the input signal S1(f) and the RBT elimination filter passed signal S2(f) (S101). There are various methods for calculating the signal level, but for example, the signal level may be calculated by integrating the square of the input signal. Hereinafter, the result of integrating the square of the input signal S1(f) will be referred to as the "signal level P1(f)," and the result of integrating the square of the RBT elimination filter passed signal S2(f) will be referred to as the "signal level P2(f)."

[0079] Next, the speech section determination unit 72 inputs the signal level P1(f) and the signal level P2(f) to an LPF (Low Pass Filter) and outputs a signal that has passed the low-frequency components in the frequency characteristics (S102). Hereinafter, the result of applying the LPF process to the signal level P1(f) will be referred to as the "signal level P1'(f)," and the result of applying the LPF process to the RBT elimination filter passed signal S2(f) will be referred to as the "signal level P2'(f)."

[0080] Note that various values ​​can be applied to the configuration and parameter values ​​of the LPF to be applied (for example, CR may be set to 8 ms (time constant)). Also, application of the LPF may be omitted.

[0081] Next, the power units of the signal levels P1'(f) and P2'(f) are converted to dBm (S103), and the signal level difference L(f) (dB) is calculated by the following equation (1) (S104). L(f) = P1'(f) - P2'(f) … (1)

[0082] Next, the process of step S105 (distinguishing between a call section and an RBT section) for determining the call section for the input signal using the signal level difference L(f) calculated above will be described.

[0083] FIG. 10 is a flowchart showing details of a call section determination process in the call section determination unit according to the second embodiment.

[0084] When the signal level difference L(f) calculated by the processing in steps S101 to S104 described above (for example, the signal level difference for each frame period of 5 ms) is input, the processing starts.

[0085] The speech section determination unit 72 determines whether or not a signal level difference of 10 dB or more has been detected within a predetermined time (e.g., 100 ms) (S105-1). For example, the speech section determination unit 72 performs this determination based on whether or not the signal level difference L(f) for each 5 ms frame period is 10 dB or more 20 times in a row (otherwise, the determination may be performed using various statistical values ​​such as the average value and the median value). If the speech section determination unit 72 detects a signal level difference of 10 dB or more within the predetermined time (e.g., 100 ms), it executes the process of the next step S105-2, whereas if it does not detect a signal level difference of 10 dB or more, it proceeds to the process of step S105-6 described later.

[0086] If a signal level difference of 10 dB or more is detected within a predetermined time by the process of step S105-1 described above, the call section determination unit 72 determines that an RBT has been detected (the start of an RBT section has been detected) (S105-2).

[0087] After detecting the RBT, the call section determination unit 72 determines whether or not a signal level difference L(f) of less than 10 dB has been detected (S105-3). The call section determination unit 72 executes the process of the next step S105-4 only when it detects a signal level difference L(f) of less than 10 dB.

[0088] Next, the speech section determination unit 72 determines whether or not a signal level difference L of 10 dB or more has not been detected for 2.5 seconds after detecting a signal level difference L(f) of less than 10 dB in the above-mentioned S105-3 (S105-4). For example, the speech section determination unit 72 determines whether or not the signal level difference L(f) for each 5 ms frame period is less than 10 dB 500 times in succession (otherwise, the determination may be made based on a statistical value such as whether or not the signal level difference L(f) is less than 10 dB on average). The speech section determination unit 72 executes the process of the next step S105-5 only when it does not detect a signal level difference L of 10 dB or more (i.e., only when it continuously detects a signal level difference L of less than 10 dB).

[0089] The speech interval determination unit 72 determines the interval from when the RBT is detected in the processing of step S105-2 described above to the point where the signal level difference L(f) of 10 dB or more is last detected as an RBT interval, and determines the remaining interval as a speech interval (S105-5).

[0090] On the other hand, if the process of step S105-1 does not detect a signal level difference L(f) of 10 dB or more within a predetermined time (100 ms), the call section determination unit 72 determines whether a predetermined time (e.g., 8 seconds) has elapsed (S105-6). The call section determination unit 72 executes the following process only if it does not detect a signal level difference L(f) of 10 dB or more within 8 seconds.

[0091] If the call section determination unit 72 does not detect a signal level difference L(f) of 10 dB or more for 8 seconds, it determines that there is no RBT section (in other words, the 8-second determination section is a call section) (S105-7).

[0092] The speech interval determination unit 72 provides the speech interval determination results (information on the speech interval) determined in the above-mentioned steps S105-5 and S105-7 to the silent interval removal and sound spurt extraction unit 73.

[0093] The silent interval removal / voice activity extraction unit 73 performs a process of removing silent intervals from the voice activity interval of the RBT removal filter passed signal S2 based on the above-mentioned determination result (information on the voice activity interval). The process of removing silent intervals from the voice activity interval removes intervals determined to be silent by the technology described in Patent Document 6 from the voice activity interval of the RBT removal filter passed signal S2, for example, and extracts a predetermined time period (e.g., 8 seconds) from the remaining voice activity interval (a section where there may be someone talking (or a section where there may be a modem signal)).

[0094] The silent section removal sound section extraction section 73 provides the signal level histogram generation section 74 with information on the extracted sound section extracted for a predetermined time period (for example, information on the signal level P2' in the extracted sound section).

[0095] FIG. 11 is a flowchart showing detailed processing of the signal level histogram generating section and the determination processing section according to the second embodiment.

[0096] The signal level histogram generating unit 74 generates a histogram of signal levels (number of occurrences for each signal level) within the extracted sound segment (S201). The signal level histogram generating unit 74 provides the histogram of signal levels within the extracted sound segment to the determination processing unit 75.

[0097] The determination processing unit 75 arranges the histogram data (extracted sound segments) provided from the signal level histogram generating unit 74 in descending order (S202).

[0098] For example, Fig. 12 is an explanatory diagram showing an example of histogram data of a voice signal and a modem signal (FAX signal) arranged in descending order. Fig. 12(A) shows an example of voice signal histogram data in an extracted sound segment, and Fig. 12(B) shows the result of arranging the top n (n value is, for example, 15) of sound segments (for example, -40 dBm or more) in Fig. 12(A) in descending order. Meanwhile, Fig. 12(C) shows an example of modem signal histogram data in an extracted sound segment, and Fig. 12(D) shows the result of arranging the top n (n value is, for example, 15) of sound segments (for example, -40 dBm or more) in Fig. 12(C) in descending order. When there are less than n pieces of histogram data indicating sound segments, less than n pieces of histogram data are used.

[0099] Based on the histogram data arranged in descending order, the determination processing unit 75 calculates the total value of the signal level (total value of the number of occurrences) within the extracted sound segment using the following formula (2) (S203). The value of n is 15, for example.

number

[0100] The determination processor 75 compares the sum sum of the number of times of the signal level in the extracted sound segment calculated by the above formula (2) with a threshold T1 (S204). If the sum sum is less than the threshold T1 (e.g., several hundred), the determination processor 75 proceeds to processing in step S209, which will be described later. On the other hand, if the sum sum is equal to or greater than the threshold T1 (e.g., several hundred), the determination processor 75 executes the following processing.

[0101] The determination processing unit 75 calculates the variance value of the signal level in the extracted sound segment by the following formula (3) based on the histogram data arranged in descending order (S205). Note that the sum in the following formula (3) is the same as the sum calculated by the above formula (2). The value of n is, for example, 15.

number

[0102] The determination processor 75 compares the variance div of the signal level within the extracted sound segment calculated by the above formula (3) with a threshold T2 (S206). If the variance div exceeds the threshold T2 (e.g., 35), the determination processor 75 determines that the signal within the extracted sound segment is speech (S207). On the other hand, if the variance div is equal to or less than the threshold T2, the determination processor 75 determines that the signal within the extracted sound segment is a signal other than speech, such as a modem, FAX, DTMF, or hold tone (S208).

[0103] Furthermore, if the sum sum is determined to be less than the threshold value T1 in the process of step S204 described above, the determination processing unit 75 determines that the signal within the extracted sound segment is silent (or the sound segment is short) (S209).

[0104] The determination processing unit 75 sends the results of the determinations made in the above-mentioned steps S207 to S209 to the audio recording data storage unit 83, and ends the process.

[0105] (B-3) Effects of the Second Embodiment According to the second embodiment, in addition to the effects of the first embodiment, the following effects can be achieved.

[0106] Based on the characteristic that a voice signal has large level fluctuations, the voice detection device 70 of the second embodiment judges whether an input signal is voice or not by referring to the variance value of a histogram generated by extracting a sound section. This allows for more accurate judgment of voice calls.

[0107] (C) Other embodiments Although various modified embodiments have been mentioned in the above-described respective embodiments, the present invention can also be applied to the following modified embodiments.

[0108] (C-1) As a modified example, in order to reduce the risk of information leakage from the recorded data, the recorded data storage unit 83 may mutes a specific signal and record the recorded data in the recording unit 84 (or the muting process may be performed before storing in the primary recording unit 81). For example, since a PB signal (push button signal) may contain credit card number or PIN number information, the PB signal is muted and recorded. A specific example of muting a PB signal is shown below.

[0109] FIG. 13 is an explanatory diagram showing a specific example of silencing a PB signal including a PB signal silencing device according to a modified embodiment.

[0110] In FIG. 13(A), PB signal muting device 300 has a PB signal detector 310 for identifying and detecting a PB signal from an input signal, a delay unit 320 for giving the input signal a delay equivalent to the detection time of the PB signal in order to synchronize with PB signal detector 310, and a PB signal muting processing unit 330 for controlling the muting of the input signal (PB signal) when a PB signal is detected.

[0111] 13B shows an example in which PB signal detector 310 performs PB detection processing on the input signal (for example, detection time is 40 ms), and if the PB signal is detected, muting control is performed on the input signal with a delay (for example, a delay of 40 ms) (i.e., there is no output (zero) of the input signal that is a PB signal). On the other hand, if the PB signal cannot be detected, the input signal is an audio signal, so the input signal is output as is.

[0112] Next, a specific configuration of the PB signal detector 310 will be described.

[0113] Since the PB signal is composed of two frequencies, combining any frequency in the low group (697, 770, 852, 941 Hz) with any frequency in the high group (1209, 1336, 1477, 1633 Hz), the PB signal detector 310 is configured to first perform band rejection of the high group or low group, and then detect the single frequencies of the low and high groups.

[0114] FIG. 14 is a configuration diagram showing an example of a PB signal detector according to a modified embodiment.

[0115] In FIG. 14, the PB signal detector 310 includes a high-group band-rejection BSF 311 that blocks the passage of the high-group band for the input signal, a low-group band-rejection BSF 312 that blocks the passage of the low-group band for the input signal, BPFs 313 that pass only the low-group bands for the input signal that has passed through the high-group band-rejection BSF 311, BPFs 314 that pass only the high-group bands for the input signal that has passed through the low-group band-rejection BSF 312, determiners 315 that determine whether the signal is a single frequency signal (697, ... 1633 Hz), and a timer processing unit 316 that monitors whether the signal continues for a certain period of time or is interrupted in order to prevent erroneous detection or momentary interruption.

[0116] The terms "S+N" and "S" in Fig. 14 are the same as those explained in Fig. 3. In addition, the method in which each determiner 315 determines whether the signal is a single-frequency signal of the high group or the low group can be, for example, the method shown in the above-mentioned determiner 120.

[0117] When each determiner 315 detects (determines) any one of the single frequency signals in the low group and any one of the single frequency signals in the high group, it means that the PB signal has been detected.

[0118] (C-2) In addition, in some call centers and the like, it is possible to input information using a dial pulse signal instead of the above-mentioned PB signal, so the dial pulse signal may also be muted and recorded in the same manner as the PB signal.

[0119] Fig. 15 is a block diagram showing an example of a dial pulse signal silencing device according to a modified embodiment. In Fig. 15, the dial pulse silencing device 400 has a dial pulse signal detector 410 for identifying and detecting a dial pulse signal from an input signal, a delay unit 420 for delaying the input signal by the detection time of the dial pulse signal in order to synchronize with the dial pulse signal detector 410, and a dial pulse signal silencing processing unit 430 for controlling silencing of the input signal (dial pulse signal) when the dial pulse signal is detected.

[0120] The configuration and dial pulse signal detection method applied to the dial pulse signal detector 410 are not particularly limited, but for example, the technology described in Patent Document 7 can be applied.

[0121] (C-3) As a modification, the recording device 80 of the first embodiment may include the voice detection device 70 of the second embodiment. That is, the recording device 80 may use the frequency detection device 100 to determine whether the signal is a modem signal, and may use the voice detection device 70 to determine whether the signal is a voice signal. The order in which the modem signal determination and the voice signal determination are applied is not particularly limited. For example, when an input signal that has been determined to be a modem signal is determined to not be a modem signal, the recording device 80 may subsequently use the voice detection device 70 to determine whether the input signal is a voice signal or not.

[0122] (C-4) In the second embodiment described above, the process of removing the RBT section was performed and then histogram data for the sound section was generated. However, if the input signal does not contain RBT, such as in a non-telephone audio conferencing service, the process of removing the RBT section may be omitted.

[0123] (C-5) In the second embodiment described above, the silent interval removal / voice activity extraction unit 73 removes silent intervals from the call interval before the signal level histogram generation unit 74 generates histogram data. However, after the histogram data is generated, signal levels that can be estimated as silent (for example, −40 dBm or less) may be excluded from the calculation, and the sum and variance of the signal levels of the remaining voice activity units may be calculated to determine whether the input signal is voice or not.

[0124] (C-6) In the second embodiment described above, if the sum sum is less than the threshold value T1 in the process of step S204, it is determined that there is no sound (or the sound period is short) and the process is terminated. However, the process of step S205 and thereafter may be executed. [Explanation of symbols]

[0125] 10, 60...communication terminal, 20...exchange, 30...gateway device, 40...switch, 50...IP network, 70...voice detection device, 71...RBT removal filter unit, 72...call section determination unit, 73...silent section removal and voice activity extraction unit, 74...signal level histogram generation unit, 75...determination processing unit, 80, 80A...recording device, 81...primary recording unit, 82...modem call / voice call determination unit, 82A...modem call / voice call determination unit, 83...recorded data storage unit, 84...recording unit, 10 0...frequency detection device, 110...BPF, 120...determinator, 300...PB signal muting device, 310...PB signal detector, 313, 314...BPF, 315...determinator, 316...timer processing unit, 320...delay unit, 330...PB signal muting processing unit, 400...dial pulse muting device, 410...dial pulse signal detector, 420...delay unit, 430...dial pulse signal muting processing unit, 311...high group band rejection BSF, 312...low group band rejection BSF.< / sn> ​< / sn> ​​​

Claims

1. a packet copy acquiring means for acquiring a packet copy of a communication packet transmitted and received between communication terminals via a network; a determination means for determining whether the communication sound is a modem call or a voice call based on the communication sound included in the duplicate packet; a storage means for sequentially recording the determined communication sound as recording data in a recording unit when the determination result of the determination means is the modem call, The determining means determines whether the communication sound is the modem call by detecting a single frequency signal specific to a modem signal.

1. A recording device comprising:

2. A duplicate packet acquisition means for acquiring a duplicate packet of a communication packet transmitted and received between communication terminals from a network; a determination means for determining whether the communication sound is a modem call or a voice call based on the communication sound included in the duplicate packet; a storage means for sequentially recording the determined communication sound as recording data in a recording unit when the determination result of the determination means is the voice call, The determination means generates histogram data regarding signal levels within a sound interval of the communication sound, arranges a predetermined number of sound intervals of the histogram data in descending order, and determines whether the communication sound is the voice call based on whether a total value of the number of times of the signal levels from the highest signal level to the predetermined highest is equal to or greater than a first threshold value and whether a variance value of the histogram data arranged in descending order of the signal levels from the highest signal level to the predetermined highest exceeds a second threshold value.

1. A recording device comprising:

3. The recording device as described in Claim 2, characterized in that the determination means uses less than the predetermined number of histogram data when the histogram data indicating the presence of sound is less than the top predetermined number.

4. The recording device according to any one of claims 1 to 3, characterized in that, when the communication sound is a push button signal or a dial pulse signal, the storage means performs a muting process on the communication sound before recording the communication sound in the recording unit.

5. Computer a packet copy acquiring means for acquiring a packet copy of a communication packet transmitted and received between communication terminals via a network; a determination means for determining whether the communication sound is a modem call or a voice call based on the communication sound included in the duplicate packet; When the determination result of the determination means is the modem call, a recording unit is caused to function as a storage means for sequentially recording the determined communication sound as recording data; The determining means determines whether the communication sound is the modem call by detecting a single frequency signal specific to a modem signal. A recording program comprising:

6. A computer a packet copy acquiring means for acquiring a packet copy of a communication packet transmitted and received between communication terminals via a network; a determination means for determining whether the communication sound is a modem call or a voice call based on the communication sound included in the duplicate packet; When the determination result of the determination means is the voice call, a recording unit is caused to function as a storage means for sequentially recording the determined communication sound as recording data; The determination means generates histogram data regarding signal levels within a sound interval of the communication sound, arranges a predetermined number of sound intervals of the histogram data in descending order, and determines whether the communication sound is the voice call based on whether a total value of the number of times of the signal levels from the highest signal level to the predetermined highest is equal to or greater than a first threshold value and whether a variance value of the histogram data arranged in descending order of the signal levels from the highest signal level to the predetermined highest exceeds a second threshold value. A recording program comprising:

7. 1. A method for recording for use in a recording device, comprising: The duplicate packet acquisition means acquires a duplicate packet of a communication packet transmitted and received between the communication terminals from the network, The determining means determines whether the communication sound included in the duplicate packet is a modem call or a voice call, the storage means, when the determination result of the determination means is the modem call, sequentially records the determined communication sound as recording data in a recording unit; The determining means determines whether the communication sound is the modem call by detecting a single frequency signal specific to a modem signal. A recording method comprising:

8. A recording method for use in a recording device, comprising: The duplicate packet acquisition means acquires a duplicate packet of a communication packet transmitted and received between the communication terminals from the network, The determining means determines whether the communication sound included in the duplicate packet is a modem call or a voice call, the storage means, when the determination result of the determination means is the voice call, sequentially records the determined communication sound as recording data in a recording unit; The determination means generates histogram data regarding signal levels within a sound interval of the communication sound, arranges a predetermined number of sound intervals of the histogram data in descending order, and determines whether the communication sound is the voice call based on whether a total value of the number of times of the signal levels from the highest signal level to the predetermined highest is equal to or greater than a first threshold value and whether a variance value of the histogram data arranged in descending order of the signal levels from the highest signal level to the predetermined highest exceeds a second threshold value. A recording method comprising:

Images