JP5419413B2 - Transmission / reception device, transmission / reception method, and program - Google Patents

Description

  The present invention relates to a transmission / reception device, a transmission / reception method, and a program for receiving a sound / acoustic signal by a transmission / reception device and transmitting the received sound / acoustic signal from the transmission / reception device to the outside.

In digital broadcasting, modulated information is transmitted from a key station to a user's tuner via a plurality of relay stations. Since various noises may be mixed in the communication path during information transmission, an error correction technique or the like is employed to reduce this influence (see, for example, Patent Documents 1 and 2). In addition, M / Stereo technology using correlation between channels is adopted as a high-efficiency encoding method such as MP3 (MPeg-1 audio layer 3), AAC (Advanced Audio Codec) or dolby-E (registered trademark). ing. (For example, see Non-Patent Documents 1 and 2)
Japanese Patent Laid-Open No. 9-18507 JP-A-1-228333 ISO / IEC 11172-3 ISO / IEC 13818-7

  However, there is a possibility that information transmission may be affected by various noises such as random noise or burst noise. Not only video data but also audio data may be affected by noise or the like in the transmission path, and this must be detected with simple processing and effectively. Also, during transmission, encoding and decoding processes are performed in various formats such as dolby-E. Quantization noise or calculation errors resulting from the encoding and decoding processes are erroneously detected as transmission fault noise. There was also a problem. Note that Patent Documents 1 and 2 and Non-Patent Documents 1 and 2 do not describe means for solving the problem.

  The present invention has been made in view of such circumstances, and an object of the present invention is to remove components accompanying encoding and decoding processes and add metadata to the audio-acoustic signal, thereby transmitting information with higher accuracy. It is an object to provide a transmission / reception apparatus, a transmission / reception method, and a program capable of monitoring noise.

Transceiver disclosed gun receives a speech acoustic signal, the receiving apparatus for transmitting audio acoustic signals received, extracting a first audio acoustic signal and the second audio acoustic signal according to the received audio acoustic signals extracted A high-pass filter that extracts a frequency component of a predetermined frequency or higher from the first audio-acoustic signal and the second audio-acoustic signal extracted by the extraction unit, and a first audio-acoustic signal related to the frequency component extracted by the high-pass filter And an addition unit that calculates a value related to a time-series sum signal of the second audio-acoustic signal, calculates addition data based on a cumulative addition value for a predetermined time of the calculated value, and a frequency component extracted by the high-pass filter. A value related to a time-series difference signal between the first audio acoustic signal and the second audio acoustic signal is calculated, and subtraction data based on a cumulative addition value for a predetermined time of the calculated value. A first subtraction unit for calculating a time-series first effective value of the first audio-acoustic signal related to the frequency component extracted by the high-pass filter, and a frequency extracted by the high-pass filter A second effective value calculation unit for calculating a second effective value in time series of the second audio-acoustic signal related to the component; metadata based on the addition data and subtraction data; metadata based on the first effective value and the second effective value Alternatively, the selection unit accepts selection of metadata based on the addition data and subtraction data and metadata based on the first effective value and the second effective value, and the selection unit accepts the selection based on the addition data and the subtraction data. When data is selected, the addition data and subtraction data calculated by the addition unit and subtraction unit are added to the received audio-acoustic signal as metadata. The first effective value and the second effective value calculated by the first effective value calculation unit and the second effective value calculation unit when the selection unit selects metadata based on the first effective value and the second effective value. A value is added to the audio-acoustic signal received as metadata, and the addition is performed when the selection unit selects metadata based on the addition data and subtraction data and metadata based on the first effective value and the second effective value. The addition data and the subtraction data calculated by the subtraction unit and the subtraction unit, and the first effective value and the second effective value calculated by the first effective value calculation unit and the second effective value calculation unit are added to the audio-acoustic signal received as metadata. And an adder for transmitting the audio-acoustic signal to which the metadata is added by the adder to the outside.

  According to the device disclosed in the present application, the extraction unit extracts the first audio acoustic signal and the second audio acoustic signal related to the received audio acoustic signal. The high-pass filter extracts a frequency component of a predetermined frequency or higher from the extracted first audio sound signal and second audio sound signal. The adding unit calculates a value related to the time-series sum signal of the first audio acoustic signal and the second audio acoustic signal related to the extracted frequency component, and adds the addition data based on a cumulative addition value for a predetermined time of the calculated value. calculate. Similarly, the subtraction unit calculates a value related to the time-series difference signal between the first audio acoustic signal and the second audio acoustic signal related to the extracted frequency component, and subtracts the calculated value based on a cumulative addition value for a predetermined time. Calculate the data. The addition data and subtraction data are added as metadata to the audio-acoustic signal by the adding unit. Finally, the transmission unit transmits the audio / acoustic signal to which the metadata is added to the outside.

  According to the device disclosed in the present application, the conversion unit converts the calculated addition data and subtraction data based on the predetermined lower limit value and upper limit value. Then, the adding unit adds the addition data and subtraction data converted by the conversion unit to the received audio-acoustic signal as metadata.

  According to the device disclosed in the present application, the extraction unit extracts the first audio acoustic signal and the second audio acoustic signal related to the received audio acoustic signal. The high-pass filter extracts a frequency component of a predetermined frequency or higher from the extracted first audio sound signal and second audio sound signal. The first effective value calculation unit calculates a first effective value in a time series of the first audio-acoustic signal related to the frequency component extracted by the high-pass filter. Similarly, a 2nd effective value calculation part calculates the 2nd time effective 2nd effective value of the 2nd audio | voice sound signal which concerns on the frequency component extracted by the high pass filter. The adding unit adds the calculated first effective value and second effective value to the received audio-acoustic signal as metadata. Finally, the transmission unit transmits the audio / acoustic signal to which the metadata is added to the outside.

  According to the device disclosed in the present application, the conversion unit performs conversion based on the logarithm of the first effective value calculated by the first effective value calculation unit and the second effective value calculated by the second effective value calculation unit. The adding unit adds the first effective value and the second effective value relating to the logarithm converted by the converting unit to the received audio-acoustic signal as metadata.

  According to one aspect of the apparatus, the high-pass filter can eliminate the direct current components of the first audio sound signal and the second audio sound signal. Therefore, when calculating addition data and subtraction data, regardless of the encoding / decoding format in which a large amount of DC component is present or the encoding / decoding format in which the DC component is excluded, in the encoding and decoding processes during transmission, The present invention has an excellent effect, for example, by making it possible to avoid a situation in which misalignment noise or the like is erroneously determined as transmission failure noise.

Embodiment 1
Embodiments of the present invention will be described below with reference to the drawings. FIG. 1 is a schematic diagram showing an outline of a transmission system. The transmission system includes a transmission / reception device 1 provided in a key station, a transmission / reception device 1 provided in a relay station, and a transmission / reception device 1. Program material (hereinafter referred to as audio data) made up of the produced video data and audio sound signals is transmitted to the key station via a plurality of relay stations. Thereafter, the video data and audio data are processed at the broadcasting station, and transmitted from the key station to a user tuner (not shown) via a plurality of relay stations. The transmission / reception device 1 provided in the key station and the relay station analyzes the audio data in the broadcast data, and calculates addition data and subtraction data (hereinafter, collectively referred to as metadata in some cases), which are feature amounts of the audio data. The transmission / reception device 1 (hereinafter, metadata calculator 1) calculates metadata from the received audio data. The metadata calculator 1 adds the calculated metadata to the audio data, and transmits it to the metadata calculator 1 of the subsequent relay station. In the following, the metadata calculator 1 that transmits audio data is the front stage, and the metadata calculator 1 that receives the audio data from the metadata calculator 1 is the rear stage.

  FIG. 2 is a block diagram showing a hardware configuration of the metadata calculator 1. The metadata calculator 1 includes a demultiplexer 11, an extraction unit 12, a metadata holding unit 13, a metadata calculation unit 14, a metadata addition unit 15, an addition unit 17, a transmission unit 18, and a high-pass filter (hereinafter referred to as HPF) 110. And the conversion unit 111 and the like. An AV stream compressed by MPEG (Moving Pictures Experts Group), AAC, dolby-E or the like is input to the metadata calculator 1. The AV stream includes both video data and audio data, but the description of the video data is omitted in this embodiment. The audio data is encoded in the AAC or dolby-E format or the like, and the audio data decoded by a decoder (not shown) and specific data (identification information) described later are input to the metadata calculator 1. In addition, when the produced video data and audio data are transmitted to a key station (broadcast station), the audio data may be transmitted without being compressed.

  The audio data and specific data input to the metadata calculator 1 are input to the demultiplexer 11. The demultiplexer 11 extracts the metadata and specific data added to the audio data, and separates and outputs the extracted metadata and specific data and the audio data from which the metadata and specific data have been removed. Note that the added metadata is metadata calculated by the metadata calculator 1 of the preceding key station or relay station. The metadata calculation process and the contents of the specific data will be described later. The audio data separated by the demultiplexer 11 is output to the adding unit 17 and the extracting unit 12, respectively. The metadata and specific data separated by the demultiplexer 11 are output to the metadata holding unit 13.

  The extraction unit 12 extracts first sound data (hereinafter, left sound data) and second sound data (hereinafter, right sound data) related to the input sound data, and the left sound data and the right sound data are passed through the HPF 110. To the metadata calculation unit 14. That is, the extraction unit 12 extracts left audio data and right audio data, respectively, when the audio data is composed of left and right 2ch, and calculates metadata of the extracted left audio data and right audio data via the HPF 110. To the unit 14.

  The extraction unit 12 performs the above-described processing when the audio data is 2ch. However, when the audio data is 3ch or more exceeding 2ch, the extraction unit 12 converts the audio data composed of three or more channels into left audio data and Conversion (downmix) into 2ch audio data composed of right audio data. The output unit 122 outputs the left audio data and the right audio data according to 2ch after conversion to the HPF 110. The conversion unit 121 stores mathematical formulas for converting audio data of 3ch or more into audio data of 2ch, and performs conversion according to the mathematical formulas. In the present embodiment, an example in which audio data is, for example, 5.1 ch will be described.

  When the input audio data is left audio data L, right audio data R, center audio data C, left surround data Ls, and right surround data Rs, the converted left audio data L ′ and the converted right audio The audio data R ′ can be expressed by the following formula (1) according to ISO / IEC 13818-7.

  Further, equation (1) does not include .1ch low-frequency effect data LFE. However, when LFE is present, left audio data L ′ after conversion and right audio data after conversion according to equation (2) below. R ′ may be calculated.

  FIG. 3 is a table showing the value of the coefficient A. This value is based on the description of 8.3.7.5 of ISO / IEC 13818-7, and the value of A is determined by the value of matrix_mixdown_idx. The table shown in FIG. 3 is also stored in the conversion unit 121. As described above, the audio data converted into the left audio data and the right audio data by the conversion unit 121 is output to the HPF 110 via the output unit 122. In the present embodiment, an example of 5.1 ch has been described, but audio data such as 7.1 ch may be converted. In addition, using the example of ISO / IEC 13818-7, an example was given in which 5.1ch audio data was downmixed to 2ch audio data composed of left audio data and right audio data. It is not limited. For example, other downmix formulas defined by ARIB / STD-B21 or the like may be used.

  The HPF 110 extracts frequency components of the left audio data and the right audio data output from the output unit 122 that are equal to or higher than a predetermined frequency. The HPF 110 outputs the left audio data and the right audio data after extracting frequency components of a predetermined frequency or higher to the metadata calculation unit 14. The predetermined frequency in the HPF 110 is stored as a cutoff frequency in an internal memory (not shown). This cut-off frequency can be appropriately changed by an input unit (not shown). The cut-off frequency may be a value of 20 Hz or less, for example, preferably about 6 Hz.

FIG. 4 is a block diagram showing the configuration of the HPF 110. The HPF 110 includes an adder 1101, an adder 1102, a delay circuit 1103, a delay circuit 1104, and the like. When the input is x [n], the output is y [n], and the intermediate output is v [n], the intermediate output v [n] can be expressed by the following equation (3).
v [n] = x [n] + a ₁ v [n-1] + a ₂ v [n-2] (3)

The output y [n] can be expressed by the following equation (4).
y [n] = b ₀ v [n] + b ₁ v [n-1] + b ₂ v [n-2] (4)

Here, the coefficient b ₀ is 0.999439161786443, the coefficient a ₁ is -1.998878015320690, the coefficient b ₁ is -1.998878323572890, the coefficient a ₂ is 0.998878631825079, and the coefficient b ₂ is 0.999439161786443. The configuration of the HPF 110 shown in FIG. 4 is an example, and other filters may be applied. For example, the HPF 110 shown in FIG. The right audio data and the left audio data from which the DC component has been removed by the HPF 110 are output to the metadata calculation unit 14.

  The metadata calculation unit 14 includes an addition unit 141 and a subtraction unit 142. The adder 141 calculates a value related to the time series sum signal of the left audio data and the right audio data, and calculates addition data based on a cumulative addition value for a predetermined time of the calculated value. The subtracting unit 142 calculates a value related to a time-series difference signal between the left audio data and the right audio data, and calculates subtraction data based on a cumulative addition value for a predetermined time of the calculated value. The calculated addition data and subtraction data are output as metadata to the metadata adding unit 15 via the conversion unit 111. Details will be described below.

  FIG. 5 is a graph schematically showing temporal changes of the left audio data and the right audio data. FIG. 5A is a graph schematically showing temporal changes in the amplitude of left audio data, and FIG. 5B is a graph schematically showing temporal changes in the amplitude of right audio data. In either case, the horizontal axis represents time, and the vertical axis represents amplitude. Input audio data is an integer multiple of 33.3ms, which is the time of one frame of NTSC (National Television Standards Committee) video, or an integral multiple of 42.6ms, which is the encoding time of one frame of AAC. ). Hereinafter, one unit of the predetermined time is referred to as a frame. In the example of FIG. 5, the audio data is divided into frame 1, frame 2,... Frame j, and addition data and subtraction data are calculated for each frame.

  In the frame 1, the left audio data can be expressed as Li, Li + 1,... Ln in chronological order according to the sampling frequency. Similarly, the right audio data in frame 1 can be expressed as Ri, Ri + 1,... Rn in time series order. The adder 141 adds the data of the left audio data at a specific time and the data of the right audio data at a specific time to calculate a sum signal. For example, the sum signal of Ri and Li is calculated. Next, the adding unit 141 calculates a value related to the sum signal by dividing the added value by 2. That is, the average value of the left audio data and the right audio data at the specific time is calculated. The adder 141 performs the processing on all time data existing in one frame. That is, arithmetic processing is performed on all time series combinations from i to n. Then, the adding unit 141 calculates the sum of the average values in the combination of all the left audio data and right audio data existing in the frame. The adding unit 141 calculates the average value of the sum by dividing the sum by the number of combinations of the left audio data and the right audio data in the frame. Specifically, the addition data AII (1) in frame 1 can be expressed by equation (5). Hereinafter, the added data is sometimes referred to as AII (Audio in-phase information).

  As described above, by using the average value in the specific time and the average value of the sum total in the frame, the added data becomes a value less than or equal to the maximum amplitude of the audio data, so that the data amount can be reduced. . In the present embodiment, the average value of the sum of the right audio data and the left audio data is calculated with respect to the addition data, but the addition value may be used without calculating the average value. That is, 1/2 of equation (5) may be calculated instead of 1 to obtain a value related to the sum signal. In this case, the addition value is calculated for all combinations of left audio data and right audio data existing in one frame. A total sum of the added values may be calculated, and an average value of the total sum may be calculated. Furthermore, although an example in which the average value of the sum is finally calculated regarding the addition data will be described, the sum may be calculated as addition data without calculating the average value. That is, an operation is performed in which 1 / n in the equation (5) is 1. The adder 141 performs the same processing for all frames 1 to j to calculate addition data AII (1) to addition data AII (j).

  Next, the subtraction unit 142 will be described. The subtracting unit 142 subtracts the data at the specific time of the right audio data from the data at the specific time of the left audio data to calculate a difference signal. Note that the subtracting unit 142 may subtract data at a specific time of the left audio data from data at a specific time of the right audio data. Next, the subtraction unit 142 calculates a value related to the difference signal by dividing the subtraction value by 2. That is, the subtraction unit 142 calculates an average value of the subtraction values at the specific time. The subtracting unit 142 performs the processing for all combinations of the left audio data and the right audio data existing in one frame. Then, the subtracting unit 142 calculates the sum of the average values, and further calculates the average value of the sums. Specifically, the subtraction data AOI (1) in frame 1 can be expressed by equation (6). Hereinafter, the subtraction data is sometimes referred to as AOI (Audio out of phase information).

  Similarly to the adding unit 141, the subtracting unit 142 may not necessarily calculate the average value of the subtracted values and the average value of the sum for the subtracted data. That is, in the present embodiment, the average value of the difference between the right audio data and the left audio data is calculated for the subtraction data, but the difference value is used as the value for the difference signal without calculating the average value. May be. That is, the calculation is performed by replacing 1/2 of Equation (6) with 1. In this case, the difference value is calculated for all combinations of left audio data and right audio data existing in one frame. The sum of the difference values may be calculated, and an average value of the sum may be calculated. Furthermore, although an example in which the average value of the sum is finally calculated regarding the subtraction data will be described, the sum may be calculated as subtraction data without calculating the average value. That is, an operation is performed in which 1 / n of the equation (6) is 1. The subtraction unit 142 performs the same processing for all the frames 1 to j to calculate subtraction data AOI (1) to subtraction data AOI (j). The adding unit 141 and the subtracting unit 142 use the addition data and the subtraction data group that have been calculated for all the frames as metadata based on the equations (5) and (6) stored in advance as metadata. To the metadata adding unit 15.

  Next, the effect of applying the HPF 110 to the previous stage of the metadata calculation unit 14 will be examined. FIG. 6 is a graph showing temporal changes in the added data when the process by the HPF 110 is not performed. In the graph of FIG. 6, the horizontal axis indicates the number of frames, and the vertical axis indicates the value of the addition data AII. A solid line indicates a temporal change of the addition data with respect to the audio data that has not been encoded and decoded. The dotted line shows the temporal change of the added data with respect to the audio data that has been encoded and decoded by dolby-E. The voice used in the experiment is about 20 seconds of female announcement, and is not affected by various noises such as random noise or burst noise.

  When comparing the solid line and dotted line graphs in FIG. 6, it can be understood that dolby-E (dotted line) is entirely offset with respect to the solid line that has not undergone encoding and decoding processing. Although not shown, it has been confirmed that the characteristics of the voice data that has undergone the encoding and decoding processes by AAC substantially match the characteristics of the voice data that has not undergone the encoding and decoding processes indicated by the solid lines. When audio data is transmitted, various types of encoding and decoding processes are applied. In that case, the added data that has undergone encoding and decoding processing such as dolby-E is different from the additional data related to the audio data that has not undergone encoding and decoding processing. On the other hand, the addition data that has undergone encoding and decoding processing such as AAC approximates the characteristics of the addition data related to audio data that has not undergone encoding and decoding processing.

  The applicant of the present application has conducted extensive research to flexibly utilize the added data related to the audio data that has undergone various types of encoding and decoding processes. As a result, by providing the HPF 110 in the preceding stage of the metadata calculation unit 14, Solved the problem. FIG. 7 is a graph showing a temporal change in the added data when the process by the HPF 110 is performed. The values on the horizontal axis and the vertical axis are the same as in FIG. The solid line shows the temporal change of the added data with respect to the audio data that has not been encoded and decoded but is processed by the HPF 110. A dotted line indicates a temporal change of the addition data with respect to the audio data subjected to the encoding and decoding processing by dolby-E and the processing by HPF 110. 7 shows an example in which a filter in which HPFs 110 shown in FIG. 4 are cascaded in four stages is used and the cutoff frequency is 6 Hz.

  As shown in FIG. 7, it can be understood that the offset amount is reduced by the frequency component extraction processing by the HPF 110, and the added data related to the audio data that has not undergone the encoding and decoding processing and the characteristics are almost the same. As a result, even when audio data that has undergone various encoding and decoding processes is received, the amount of offset can be reduced, so that it is possible to avoid a situation in which quantization noise or the like is erroneously detected as transmission failure noise. The cut-off frequency used in the experiment is 6 Hz, but a value of 20 Hz or less may be adopted as appropriate. This is because a frequency of 20 Hz or less is a frequency with low hearing sensitivity for humans, and even if a fault occurs in this frequency band, it cannot be heard, that is, humans cannot recognize it as a fault.

  The effect of providing the HPF 110 in the previous stage of the metadata calculation unit 14 has been demonstrated also in the subtraction data AOI. FIG. 8 is a graph showing temporal changes in subtraction data when the process by the HPF 110 is not performed. In the graph of FIG. 8, the horizontal axis indicates the number of frames, and the vertical axis indicates the value of the subtraction data AOI. A solid line indicates a temporal change of subtraction data with respect to audio data that has not been encoded and decoded. A dotted line indicates a temporal change of subtraction data with respect to audio data that has been encoded and decoded by dolby-E. As shown in FIG. 8, the subtraction data that has undergone the encoding and decoding processing by dolby-E is greatly different from the subtraction data that has not undergone the encoding and decoding processing. Similarly to the addition data, the subtraction data differs in offset amount depending on the format of encoding and decoding processing.

  FIG. 9 is a graph showing temporal changes in the subtraction data when the processing by the HPF 110 is performed. The values on the horizontal and vertical axes are the same as in FIG. The solid line shows the temporal change of the subtraction data with respect to the audio data that has not been encoded and decoded, but has been processed by the HPF 110. A dotted line shows a temporal change of subtraction data with respect to audio data that has been encoded and decoded by dolby-E and processed by HPF 110. The configuration and cutoff frequency of the HPF 110 were the same as those used in the addition data experiment. As shown in FIG. 9, it can be understood that the subtraction data in the subtraction data is substantially the same as the subtraction data in which the offset amount is reduced and the encoding and decoding processes are not performed, as in the addition data.

  The addition data and subtraction data calculated by the metadata calculation unit 14 are output to the conversion unit 111. The conversion unit 111 includes an upper / lower limit conversion unit 1110 and an integer conversion unit 1111. An upper / lower limit conversion unit 1110 obtains absolute values of addition data and subtraction data, and converts the absolute values of addition data and subtraction data based on an upper limit value and a lower limit value stored in an internal memory (not shown). Specifically, 3 is stored in the memory as the lower limit value and 255 as the upper limit value. In addition, these numerical values are examples and are not limited thereto. When the absolute values of the addition data and the subtraction data are smaller than the lower limit value 3, the upper / lower limit conversion unit 1110 converts the values to zero. It is not always necessary to convert to zero, and it may be converted to 1 or 2 smaller than 3. In this case, the upper / lower limit conversion unit 1110 adds the code before the absolute value calculation, which was previously assigned to the addition data and subtraction data before conversion, to the addition data and subtraction data after conversion. For example, when the addition data is −2, the absolute value 2 is calculated and converted to 1 by the conversion process. Finally, the upper and lower limit conversion unit 1110 adds the original code-to obtain -1.

  Furthermore, when the absolute values of the addition data and the subtraction data exceed the upper limit value 255, the upper / lower limit conversion unit 1110 converts the absolute values of the addition data and the subtraction data into a value less than the upper limit value 255 or 255. Next, the upper and lower limit conversion unit 1110 adds the code before the absolute value calculation, which is previously given to the addition data and subtraction data before conversion, to the addition data and subtraction data after conversion. Specifically, when the addition data is −329, the absolute value of the addition data is calculated as 329, which exceeds the upper limit value. The upper / lower limit conversion unit 1110 converts the absolute value of the addition data into an upper limit value 255 or 73 (329-256) less than the upper limit value. When converting to less than the upper limit value, for example, 254 or 253 stored in the memory in advance may be used, or zero may be used. In addition, as an example of processing for conversion to less than the upper limit value, a value related to the nth power of 2 is subtracted from the absolute value of the addition data or subtraction data, and converted to a numerical value that can be expressed by a predetermined bit (for example, 8 bits) or less Also good. In this example, the value related to the power of 2 is subtracted. In addition, a value related to 2 to the seventh power may be subtracted.

  The upper / lower limit conversion unit 1110 adds the code before the absolute value calculation, which has been previously given to the addition data and subtraction data before conversion, to the addition data and subtraction data after conversion. In the above example, since the addition data is −329, a sign − is added to 73 after conversion to obtain −73. When the addition data is smaller, for example, −956, the upper limit value 255 or less than the upper limit value 188 (956-512-256) is converted. In this example, 2 9 to the 2nd power and 2 to the 8th power are subtracted as values related to the 2 n power so that the converted value is 8 bits or less. Next, the upper and lower limit conversion unit 1110 adds the code before the absolute value calculation, which is previously given to the addition data and subtraction data before conversion, to the addition data and subtraction data after conversion. In this example, the addition data is −188.

  In the present embodiment, the upper / lower limit conversion unit 1110 describes processing for obtaining the absolute values of the addition data and the subtraction data having positive and negative values and then converting them. However, the present invention is not limited to this. The upper / lower limit conversion unit 1110 stores a first range and a second range corresponding to the lower limit value in a memory (not shown). For example, the first range is stored as being larger than −3 and smaller than 0, and the second range is stored as being larger than 0 and smaller than +3, for example. The upper and lower limit conversion unit 1110 stores a third range and a fourth range corresponding to the upper limit value in a memory (not shown). For example, the third range is stored as being smaller than −255, and the fourth range is stored as being larger than +255.

  When receiving the addition data and the subtraction data, the upper / lower limit conversion unit 1110 determines whether the addition data and the subtraction data belong to the first range to the fourth range. When it is determined that the upper / lower limit conversion unit 1110 belongs to the first range, the addition data and the subtraction data are converted to, for example, zero or a negative value larger than −3, for example, −1. When it is determined that the upper / lower limit conversion unit 1110 belongs to the second range, the addition data and the subtraction data are converted into, for example, zero or a positive value smaller than +3, for example, +1. When it is determined that the upper / lower limit conversion unit 1110 belongs to the third range, the upper / lower limit conversion unit 1110 performs conversion to a negative value larger than −255 or −255. For example, it is possible to add −254, zero, or a value related to the nth power of 2 from addition data or subtraction data to obtain a negative numerical value that can be expressed by a predetermined bit (for example, 8 bits) or less. Similarly, when the upper / lower limit conversion unit 1110 determines that it belongs to the fourth range, it converts it to a positive value smaller than +255 or 255. For example, +254, zero, or a positive numerical value that can be expressed by a predetermined bit (for example, 8 bits) or less may be obtained by subtracting a value of 2 to the nth power from addition data or subtraction data.

  The integer conversion unit 1111 rounds off the decimal point of the addition data and the subtraction data, rounds it up, rounds it off, or the like, thereby converting the addition data and the subtraction data into integer values. In addition, when addition data and subtraction data are input to the conversion unit 111, the processing by the integer conversion unit 1111 may be performed after the conversion processing by the upper and lower limit conversion unit 1110. Conversely, the conversion processing by the upper / lower limit conversion unit 1110 may be performed after the addition data and the subtraction data are converted into integers by the integer conversion unit 1111. In the present embodiment, an example will be described in which the conversion by the upper / lower limit conversion unit 1110 is first performed and then the integer conversion unit 1111 performs conversion to an integer value. When the addition data and the subtraction data are zero, and when the absolute values of the addition data and the subtraction data are 3 or more and 255 or less, the upper / lower limit conversion unit 1110 does not perform the conversion process. In this case, the addition data and the subtraction data output from the metadata calculation unit 14 are only converted into integers by the integer conversion unit 1111 of the conversion unit 111.

  FIG. 10 is a graph showing temporal changes in the added data that has undergone each encoding and decoding process. In FIG. 10, the horizontal axis indicates the frame, and the vertical axis indicates the value of the added data. A solid line abs indicates a temporal change in a value obtained by subtracting the addition data that has undergone the encoding and decoding processing by AAC from the addition data that has not undergone the encoding and decoding processing. A dotted line abs shows a temporal change in absolute value obtained by subtracting the addition data that has undergone encoding and decoding processing by dolby-E from the addition data that has not undergone encoding and decoding processing. The voice data used in the experiment is the same as the voice data used in the description of FIGS.

  FIG. 11 is a graph showing temporal changes in subtraction data that has undergone each encoding and decoding process. In FIG. 11, the horizontal axis indicates the frame, and the vertical axis indicates the value of the subtraction data. A solid line abs indicates a temporal change in a value obtained by subtracting subtracted data that has undergone encoding and decoding processing by AAC from subtracted data that has not undergone encoding and decoding processing. A dotted line abs shows a temporal change in absolute value obtained by subtracting subtracted data that has undergone encoding and decoding processing by dolby-E from subtracted data that has not undergone encoding and decoding processing. As shown in FIGS. 10 and 11, addition data and subtraction data that have undergone various encoding and decoding processes are addition data and subtraction data that have not undergone encoding and decoding processes even if they are audio data that does not have transmission disturbance noise. On the other hand, it has a difference of about 3 or less.

  This difference is caused by various encoding and decoding processes, and may be determined as transmission failure noise. Therefore, in the present embodiment, the lower limit value 3 is stored in the memory of the upper / lower limit conversion unit 1110, and the absolute values of the addition data and the subtraction data smaller than 3 are converted to 0, whereby each encoding and decoding is performed. It is intended to reduce erroneous determination of quantization noise or the like of processed audio data as failure noise. On the other hand, the absolute values of the addition data and the subtraction data rarely exceed 255 as shown in FIGS. Therefore, in the present embodiment, the upper limit value 255 is stored in the memory of the upper / lower limit conversion unit 1110, and the absolute value of the addition data and subtraction data exceeding 255 is converted into a value less than the upper limit value 255 or 255. Therefore, the amount of information at the time of transmission is reduced.

  Further, the integer conversion unit 1111 reduces the amount of information by converting the absolute values of the addition data and the subtraction data into integers. As a result, after passing through the conversion unit 111, the addition data and the subtraction data become 0 or an integer value of 3 to 255, and the numerical value of 1 or 2 does not exist. It is converted into a total of 8-bit value of 127 7-bit value and 1-bit code value, and it is possible to reduce the amount of information, and erroneous detection of quantization noise or the like due to the type of encoding and decoding processing Can be reduced. The addition data and the subtraction data converted by the conversion unit 111 are output to the metadata adding unit 15.

  FIG. 12 is an explanatory diagram showing a record layout of the metadata holding unit 13. The metadata holding unit 13 stores the metadata and specific data calculated by the preceding metadata calculators 1, 1... Output from the demultiplexer 11. The metadata holding unit 13 includes a station ID field, a device ID feed, and a metadata field. The station ID is a unique identifier assigned in advance to the key station and the relay station. The station ID, for example, means that the smaller the numerical value, the earlier the station ID. In this example, station ID01 is a key station, meaning that there is a relay station with station ID02 in the subsequent stage, a relay station with station ID03 in the subsequent stage, and a relay station with station ID04 in the subsequent stage.

  The station ID of the relay station in this example is assumed to be 05 in the subsequent stage. The device ID is a unique identifier assigned in advance for specifying the metadata calculator 1 installed in each of the key station and the relay station. For example, a MAC (Media Access Control) address may be used as the device ID. In this embodiment, two types of station ID and device ID are described. However, any one of them may be used. Regarding the device ID, information indicating which device ID is the device ID related to the metadata calculator 1 in the previous stage is stored in a memory (not shown). The metadata calculated by the specific metadata calculator 1 is associated with the metadata calculator 1 and the station ID and device ID for specifying the metadata. Hereinafter, the station ID for identifying the calculated metadata and the device ID are referred to as specific data.

  In the metadata field, metadata calculated by the preceding metadata calculator 1 is stored. The metadata holding unit 13 stores the metadata calculated by the metadata calculators 1, 1,... In the previous stage and specific data for specifying the metadata as a history. The metadata holding unit 13 outputs the metadata and the specific data to the metadata adding unit 15 as an identification information adding unit. When the metadata calculator 1 shown in FIG. 2 exists in the key station, no data is stored in the metadata holding unit 13 because there is no preceding stage.

  The metadata adding unit 15 functioning as an identification information adding unit adds station ID (05 in this example) and specific data related to the device ID to the metadata output from the metadata calculating unit 14. Further, the metadata adding unit 15 performs processing for adding the metadata and specific data to the previous metadata and specific data output from the metadata holding unit 13.

  FIG. 13 is an explanatory diagram showing the data structure of metadata and specific data. As shown in FIG. 13, the metadata and specific data calculated by each metadata calculator 1 are combined with the header in the order of transmission. That is, the metadata is combined in order from the smallest station ID. The metadata calculated by the key station metadata calculator 1 is stored in the forefront together with the specific data of the station ID01. The metadata calculated by the metadata calculator 1 is stored in the last stage together with the specific data of the station ID 05. The metadata and specific data to which the previous history is added by the metadata adding unit 15 are output to the adding unit 17.

  The adding unit 17 adds the metadata and specific data output from the metadata adding unit 15 to the audio data output from the demultiplexer 11 and outputs the audio data to the transmitting unit 18. The transmission unit 18 transmits audio data encoded by an encoder (not shown), metadata added thereto, and specific data to the metadata calculator 1 provided in the subsequent relay station. As a result, the metadata and specific data calculated by each metadata calculator 1 are sequentially added to the audio data.

  The procedure of the metadata calculation process and the addition process in the above hardware configuration will be described using a flowchart. 14 to 16 are flowcharts showing the procedures of the metadata calculation process and the addition process. The demultiplexer 11 determines whether metadata and specific data are added to the input audio data (step S71). If the demultiplexer 11 determines that metadata and specific data are added (YES in step S71), the demultiplexer 11 extracts the metadata and specific data from the audio data (step S72).

  The demultiplexer 11 outputs the metadata and specific data to the metadata holding unit 13 (step S73). If it is determined in step S71 that metadata and specific data are not added to the audio data (NO in step S71), the processes in steps S72 and S73 are skipped. Further, the demultiplexer 11 outputs the audio data to which the metadata and the specific data are not added to the extraction unit 12 and the addition unit 17 (step S74). The extraction unit 12 determines whether or not the audio data has more than 2 channels (step S75).

  When the extraction unit 12 determines that the audio data has more than 2ch channels (YES in step S75), the conversion unit 121 reads the equation (1) and converts it into 2ch audio data by substituting numerical values. The 2ch audio data is output via the output unit 122 (step S76). After the process of step S76, the extraction unit 12 proceeds to step S77. If it is determined in step S75 that the audio data is not the number of channels exceeding 2ch (NO in step S75), that is, if it is determined that the signal is a 2ch signal, the extraction unit 12 skips the process of step S76 and the left audio Data and right audio data are extracted (step S77).

  The extraction unit 12 outputs the left audio data and the right audio data to the HPF 110 (step S78). The HPF 110 reads the cutoff frequency stored in the memory. The HPF 110 extracts a frequency component exceeding the read cutoff frequency (step S79). Specifically, the HPF 110 inputs left audio data and right audio data to the filter shown in FIG. 4 in order to reduce factors such as quantization noise of addition data and subtraction data regardless of encoding and decoding processing, respectively. And get the output. The HPF 110 outputs the extracted left audio data and right audio data to the metadata calculation unit 14 (step S710). The adder 141 reads the equation (5) and substitutes the left audio data and the right audio data into the equation (5) to calculate the addition data of each frame (step S711). The subtracting unit 142 reads the equation (6) and substitutes the left audio data and the right audio data into the equation (6) to calculate the subtraction data for each frame (step S712).

  The metadata calculation unit 14 outputs the addition data and subtraction data of each frame as metadata to the conversion unit 111 (step S713). The upper and lower limit conversion unit 1110 of the conversion unit 111 reads the lower limit value from the memory. The upper / lower limit conversion unit 1110 converts the absolute values of the addition data and subtraction data smaller than the read lower limit value to zero (step S714). The upper and lower limit conversion unit 1110 of the conversion unit 111 reads the upper limit value from the memory. The upper / lower limit conversion unit 1110 converts the absolute value of the addition data and subtraction data exceeding the read upper limit value into a value lower than the upper limit value or the upper limit value (step S715). The upper / lower limit conversion unit 1110 converts the addition data and subtraction data (upper limit value or subtraction data) converted in step S715 into the addition data and subtraction data before conversion in step S715, which are given in advance. (Value less than the upper limit value) (step S716). The integer conversion unit 1111 of the conversion unit 111 converts addition data and subtraction data having positive and negative values into integers (step S81).

  The conversion unit 111 outputs the addition data and subtraction data of each frame after conversion to the metadata addition unit 15 as metadata (step S82). The metadata adding unit 15 reads the station ID and device ID related to the metadata calculator 1 stored in a memory (not shown), and adds it to the metadata output from the metadata calculating unit 14 (step S83). The metadata adding unit 15 adds the metadata and the specific data related to the previous metadata calculator 1 output from the metadata holding unit 13 to the metadata to which the specific data is added in step S83 (step S84). For example, the metadata adding unit 15 sorts each metadata and specific data in ascending order of the numerical value of the station ID or device ID so that the station ID or device ID in the preceding stage is higher, and the metadata and specific data shown in FIG. Generate a group of data.

  The metadata adding unit 15 outputs the metadata and the specific data to the adding unit 17 (step S85). The adding unit 17 adds the metadata and specific data output from the metadata adding unit 15 to the audio data output from the demultiplexer 11 (step S86). The adding unit 17 outputs the audio data, metadata, and specific data encoded together with the video data to the transmitting unit 18 (step S87). The transmission unit 18 transmits video data, audio data, metadata, and specific data to the metadata calculator 1 at the subsequent stage (step S88).

Embodiment 2
The second embodiment relates to a mode in which the metadata calculation unit 14 calculates an effective value (RMSV: Root Mean Square Value). FIG. 17 is a block diagram showing a hardware configuration of the metadata calculator 1 according to the second embodiment. The configuration of the metadata calculation unit 14 and the conversion unit 111 is different from that of the first embodiment. The metadata calculation unit 14 includes a left effective value calculation unit 143 that is a first effective value calculation unit and a right effective value calculation unit 144 that is a second effective value calculation unit. The left effective value calculation unit 143 calculates the effective value of the left audio data output from the HPF 110 based on the equation (7), and outputs it to the conversion unit 111. The effective value (LARI) (1) of the first frame of the left audio data can be expressed by Expression (7). In the following, the effective value is sometimes referred to as ARI (Audio root-mean-square information).

  Further, the effective value (RARI) (1) of the first frame of the right audio data can be expressed by Expression (8).

The effective values of the left audio data and the right audio data of each frame calculated by the metadata calculation unit 14 are output to the conversion unit 111. The conversion unit 111 performs conversion based on the logarithm of the effective value of the input left audio data and right audio data. The conversion unit 111 includes a logarithmic conversion unit 1112 and an integer conversion unit 1113. The logarithmic conversion unit 1112 calculates the logarithm of the effective value of the left audio data and the right audio data of each frame based on the equations (9) and (10).
L = k × log ₁₀ (LARI (i)) (9)
R = k × log ₁₀ (RARI (i)) (10)

  The logarithm L of the left audio data can be obtained by multiplying the logarithm of the effective value of the left audio data of each frame by a coefficient k, as shown in Equation (9). Also, the logarithm R of the right audio data can be obtained by multiplying the logarithm of the effective value of the right audio data of each frame by the coefficient k as shown in the equation (10). The coefficient k may be 50, for example. The integer conversion unit 1113 converts the logarithm of the right audio data and the logarithm of the left audio data obtained by the equations (9) and (10) to an integer by rounding down, rounding up, or rounding. The effective value related to the logarithm of the left audio data and the right audio data after being converted by the conversion unit 111 is output to the metadata adding unit 15.

  FIG. 18 is a graph showing temporal changes in the effective value of audio data. In FIG. 18, the horizontal axis represents a frame, and the vertical axis represents the effective value of left audio data. The audio data was a female announcement recorded for a few seconds. Since the effective value takes a value from 0 to about 8000 as shown on the vertical axis, the amount of information increases during transmission. Therefore, in this embodiment, it is possible to reduce the amount of information by taking a logarithm and further converting it to an integer.

  FIG. 19 is a graph showing the logarithmic change in the effective value of the audio data. In FIG. 19, the horizontal axis represents the frame, and the vertical axis represents the coefficient of the effective value of the left audio data. In the present embodiment, the coefficient k in equation (7) is 50. As a result, the logarithm of the effective value belongs to an integer value in the range of 0 to 255, and the metadata can be stored in the 8-bit information amount as in the first embodiment. As described above, the information amount can be set to the same number of bits as in the first embodiment by performing a series of processes of logarithmic conversion, coefficient multiplication, and integer conversion.

  The coefficient k may be changed dynamically by the conversion unit 111. The coefficient k is calculated so that the maximum logarithm value of the left audio data or the maximum logarithm value of the right audio data obtained by the logarithmic conversion unit 1112 is closest to a predetermined upper limit value, for example, 255. Specifically, a value obtained by dividing the upper limit value by the maximum logarithm value of the left audio data or the logarithm value of the right audio data output from the logarithmic conversion unit 1112 may be used as the coefficient k.

  FIG. 20 is a flowchart showing the metadata calculation process and the conversion process. After the process of step S710 described in the first embodiment, the following process is executed. The left effective value calculation unit 143 calculates an effective value related to the left audio data of each frame (step S201). Similarly, the right effective value calculation unit 144 calculates an effective value related to the right audio data (step S202). The metadata calculation unit 14 outputs the effective values related to the calculated left audio data and right audio data of each frame to the conversion unit 111 (step S203). The logarithmic conversion unit 1112 of the conversion unit 111 calculates the logarithm of the effective value related to the left audio data and the right audio data (step S204).

  The logarithmic conversion unit 1112 reads a coefficient k stored in advance from a memory (not shown) (step S205). The logarithmic conversion unit 1112 multiplies the logarithm by a coefficient k (step S206). The integer conversion unit 1113 converts the logarithm related to the effective value of the left audio data and the right audio data after multiplication into an integer (step S207). Finally, the conversion unit 111 outputs the logarithms relating to the effective values of the left audio data and the right audio data of each frame converted to integers as metadata to the metadata adding unit 15 (step S208). Since the subsequent processing is the same as that after step S83, detailed description is omitted.

  The second embodiment is configured as described above, and the other configurations and operations are the same as those of the first embodiment. Therefore, corresponding parts are denoted by the same reference numerals, and detailed description thereof is omitted.

Embodiment 3
Embodiment 3 relates to a mode of transmitting metadata according to a plurality of formats. FIG. 21 is a block diagram illustrating a hardware configuration of the metadata calculator 1 according to the third embodiment. The output of the HPF 110 is provided with a switch S1 and a switch S2 that are turned on or off according to the control of the selection unit 112. The selection unit 112 is, for example, a switch or the like provided on the outer surface of the casing (not shown) of the metadata calculator 1. The selection unit 112 can alternatively select either using the addition data and the subtraction data as metadata or using the effective value as metadata. When the selection unit 112 selects the addition data and the subtraction data as metadata (hereinafter referred to as a first selection), the selection unit 112 controls the switch S1 to be turned on and the switch S2 to be turned off.

  In this case, the left audio data and the right audio data output from the HPF 110 are output to the metadata calculation unit 14 including the addition unit 141 and the subtraction unit 142 via the switch S1. Hereinafter, the metadata calculation unit 14 side including the addition unit 141 and the subtraction unit 142 is referred to as a first selection side S10. On the other hand, it is not output to the metadata calculation unit 14 including the left effective value calculation unit 143 and the right effective value calculation unit 144. Hereinafter, the metadata calculation unit 14 side including the left effective value calculation unit 143 and the right effective value calculation unit 144 is referred to as a second selection side S20.

  The addition data and the subtraction data converted by the conversion unit 111 in the first selection side S10 are output to the metadata adding unit 15. The selection unit 112 outputs information indicating the type of metadata to the metadata adding unit 15. Specifically, the “first” information is transmitted. When the selection unit 112 selects the effective value as metadata (hereinafter referred to as a second selection), the selection unit 112 controls the switch S1 to be turned off and the switch S2 to be turned on.

  In this case, the left audio data and the right audio data output from the HPF 110 are output to the metadata calculation unit 14 of the second selection side S20 via the switch S2. The logarithm of the effective value relating to the left audio data and the right audio data subjected to the conversion process by the conversion unit 111 in the second selection side S20 is output to the metadata adding unit 15. The selection unit 112 outputs information indicating the type of metadata to the metadata adding unit 15. Specifically, “second” information is transmitted.

  The metadata adding unit 15 is provided with an identifier adding unit 151. The identifier adding unit 151 newly adds a type of metadata added as specific data. FIG. 22 is an explanatory diagram showing a data structure of metadata and specific data according to the third embodiment. FIG. 22A is an explanatory diagram showing the data structure of the metadata and specific data when the first selection is made. Information indicating the type of metadata is newly added after the device ID added as the specific data. At the time of verification, the metadata type is analyzed by referring to the type of metadata described in the specific data. In the example of FIG. 22A, “first” indicating the first selection is added as the type of metadata.

  FIG. 22B is an explanatory diagram showing the data structure of metadata and specific data when the second selection is made. In the example of FIG. 22B, “second” indicating the second selection is added as the type of metadata. As shown in FIG. 22B, in the metadata calculator 1 with the station ID “01”, the metadata type “first” is added, and with the metadata calculator 1 with the station ID “05”. It can be understood that the metadata type “second” is added. The accuracy of noise analysis using metadata related to addition data and subtraction data, or noise analysis using metadata related to effective values, depends on the communication environment, device characteristics, voice type, etc. It is set as the structure which can select these alternatively. The metadata adding unit 15 outputs the metadata and specific data to the adding unit 17.

  23 and 24 are flowcharts showing the procedure of the metadata adding process according to the third embodiment. The selection unit 112 accepts the first selection or the second selection in advance (step S231). The selection unit 112 determines whether or not the first selection has been accepted (step S232). If the selection unit 112 determines that the first selection has been received (YES in step S232), the switch S1 is turned on and the switch S2 is turned off (step S233). Subsequently, the following processing from step S71 to step S79 of the first embodiment is executed.

  The HPF 110 outputs the left audio data and the right audio data to the metadata calculation unit 14 of the first selection side S10 via the switch S1 (step S234). The conversion unit 111 outputs the converted addition data and subtraction data as metadata to the metadata addition unit 15 (step S235). Note that the details of steps S234 and S235 are as described in the first embodiment, and a description thereof will be omitted. The selection unit 112 outputs information indicating the type of metadata to the metadata adding unit 15 (step S236).

  If the selection unit 112 determines that the first selection has not been received (NO in step S232), the switch S1 is turned off and the switch S2 is turned on (step S237). Subsequently, the following processing from step S71 to step S79 of the first embodiment is executed. The HPF 110 outputs the left audio data and the right audio data to the metadata calculation unit 14 on the second selection side S20 via the switch S2 (step S238). The conversion unit 111 outputs logarithms related to the effective values of the converted left audio data and right audio data to the metadata adding unit 15 as metadata (step S239). Note that the details of steps S238 and S239 are the same as described in the second embodiment, and a description thereof will be omitted. The selection unit 112 outputs information indicating the type of metadata to the metadata adding unit 15 (step S241).

  The metadata adding unit 15 reads the station ID and device ID related to the metadata calculator 1 stored in a memory (not shown). The identifier adding unit 151 adds the type of metadata output from the selection unit 112, the station ID, and the device ID as specific data to the metadata output from the metadata calculation unit 14 (step S242). The metadata adding unit 15 adds the metadata and specific data related to the previous metadata calculator 1 output from the metadata holding unit 13 to the metadata to which the specific data is added in step S242 (step S243).

  For example, the metadata adding unit 15 sorts each metadata and specific data in ascending order of the numerical value of the station ID or device ID so that the station ID or device ID in the preceding stage is higher, and the metadata and specific data shown in FIG. Generate a group of data. The metadata adding unit 15 outputs the metadata and the specific data to the adding unit 17 (step S244). Subsequent processing is the same as that after step S86 of the first embodiment, and thus detailed description thereof is omitted. As described above, since a plurality of metadata calculation units 14 are provided so that they can be selected alternatively, a more suitable metadata can be selected according to the communication environment, types of encoding and decoding processes, device characteristics, types of audio data, and the like. The data calculation unit 14 can be used.

  The third embodiment is configured as described above, and the other configurations and operations are the same as those of the first and second embodiments. Therefore, corresponding parts are denoted by the same reference numerals, and detailed description thereof is omitted. To do.

Embodiment 4
The fourth embodiment relates to a mode in which a plurality of types of metadata are added together. The selection unit 112 can select any three of the first selection side S10, the second selection side S20, or the first selection side S10 and the second selection side S20. When the first selection side S10 and the second selection side S20 are selected, the selection unit 112 performs control so that both the switch S1 and the switch S2 are turned on. In this case, the HPF 110 outputs the left audio data and the right audio data to both the metadata calculators 14 of the first selection side S10 and the second selection side S20 via the switches S1 and S2.

  The selection unit 112 outputs information indicating the type of metadata to the metadata adding unit 15. In this case, the selection unit 112 transmits “first and second” information, for example. In the present embodiment, a mode in which metadata relating to both the first selection side S10 and the second selection side S20 is added through selection by the selection unit 112 will be described, but the present invention is not limited to this. Without providing the selection unit 112, the left audio data and the right audio data may always be output from the HPF 110 to the metadata calculation unit 14 on the first selection side S10 and the metadata calculation unit 14 on the second selection side S20. .

  The identifier adding unit 151 of the metadata adding unit 15, when “first and second” is output as information indicating the type of metadata from the selection unit 112, the metadata on the first selection side S 10 and the second selection side Information indicating the type of metadata is added to each metadata of S20. FIG. 25 is an explanatory diagram showing a data structure of metadata and specific data according to the fourth embodiment. The identifier adding unit 151 adds information “first” indicating the type of metadata to the metadata output from the converting unit 111 of the first selection side S10. In the example of FIG. 25, “first” is added as the specific data of the metadata related to the addition data and the subtraction data.

  Similarly, the identifier adding unit 151 adds information “second” indicating the type of metadata to the metadata output from the conversion unit 111 of the second selection side S20. In the example of FIG. 25, “second” is added as the specific data of the metadata relating to the logarithm of the left audio data and the right audio data. The metadata adding unit 15 reads the station ID and the device ID from the memory and adds them to the metadata as specific data.

  As shown in FIG. 25, the metadata adding unit 15 includes the type and metadata of the first selection side S10 after the station ID and the device ID, and further the type of metadata of the second selection side S20. And concatenate metadata. Note that the data structure of the specific data and metadata described in the present embodiment is merely an example, and the present invention is not limited to this. The order is not limited to this as long as the metadata type and the station ID or device ID can be specified.

  26 to 28 are flowcharts showing the procedure of the metadata adding process according to the fourth embodiment. The selection unit 112 accepts either the first selection, the second selection, or the first selection and the second selection (step S261). The selection unit 112 determines whether the first selection has been accepted (step S262). If the selection unit 112 determines that the first selection has been received (YES in step S262), the switch S1 is turned on and the switch S2 is turned off (step S263). Thereafter, the processing after step S234 is executed (step S264), and the series of processing ends.

  If the selection unit 112 determines that the first selection is not received (NO in step S262), the selection unit 112 determines whether both the first selection and the second selection are received (step S265). If the selection unit 112 determines that both the first selection and the second selection are not accepted (NO in step S265), the switch S1 is turned off and the switch S2 is turned on because the second selection is accepted (step S266). . Thereafter, the processing after step S238 is executed (step S267), and the series of processing ends. If the selection unit 112 determines that both the first selection and the second selection have been received (YES in step S265), the switch S1 is turned on and the switch S2 is turned on (step S268). Subsequently, the following processing from step S71 to step S79 of the first embodiment is executed.

  The selection unit 112 outputs information indicating the type of metadata to the metadata adding unit 15 (step S269). The HPF 110 outputs the left audio data and the right audio data to the metadata calculation unit 14 of the first selection side S10 via the switch S1 (step S271). The conversion unit 111 outputs the converted addition data and subtraction data as metadata to the metadata addition unit 15 (step S272). The identifier adding unit 151 refers to the metadata type output in step S269, and adds the metadata type as specific data to the metadata (step S273). Specifically, the identifier adding unit 151 adds “first” indicating the first selection side S10 as the specific data before the metadata related to the addition data and the subtraction data.

  Further, the HPF 110 outputs the left audio data and the right audio data to the metadata calculation unit 14 on the second selection side S20 via the switch S2 (step S274). The conversion unit 111 outputs logarithms related to the effective values of the converted left audio data and right audio data as metadata to the metadata adding unit 15 (step S275). The identifier adding unit 151 refers to the metadata type output in step S269, and adds the metadata type as specific data to the metadata (step S276). Specifically, the identifier adding unit 151 adds “second” indicating the second selection side S20 as specific data to the preceding stage of the metadata relating to the logarithm of the left audio data and the right audio data.

  The identifier adding unit 151 connects the metadata type and metadata of the second selection side S20 added in step S276 to the metadata type and metadata of the first selection side S10 added in step S273 (step S277). ). The metadata adding unit 15 reads the station ID and the device ID from the memory (step S278). The metadata adding unit 15 adds the read station ID and device ID as specific data to the preceding stage of the specific data and metadata connected in step S277 (step S279). Thereby, the specific data and metadata of the station ID (05) shown in FIG. 25 are completed.

  The metadata adding unit 15 further adds the metadata and the specific data related to the metadata calculator 1 in the previous stage output from the metadata holding unit 13 to the metadata to which the specific data is added in step S279 (step S281). ). The metadata adding unit 15 outputs the metadata and the specific data to the adding unit 17 (step S282). As described above, since the amount of information is reduced by the conversion unit 111, a plurality of types of metadata can be transmitted together. As a result, it is possible to detect noise more accurately using a plurality of metadata regardless of the communication environment, the types of encoding and decoding processes, the device characteristics, the types of audio data, and the like.

  The fourth embodiment is configured as described above, and the other configurations and operations are the same as those of the first to third embodiments. Therefore, the corresponding parts are denoted by the same reference numerals and detailed description thereof is omitted. To do.

Embodiment 5
The processing according to Embodiments 1 to 4 may be realized as software processing using the computer shown in FIG. FIG. 29 is a block diagram showing a hardware configuration of the metadata calculator 1 according to the fifth embodiment. The computer 10 includes a CPU (Central Processing Unit) 101, a RAM (Random Access Memory) 102, a storage unit 105 such as a hard disk, I / Fs 106 and 108 as interfaces, a communication unit 109, and the like. The CPU 101 is connected to each hardware via the bus 107, and executes the various software processes described above according to the processing program 105P stored in the storage unit 105.

  A program for operating the computer 10 can also be provided by a portable recording medium 1A such as a CD-ROM, MO, or DVD-ROM. Furthermore, the program can be downloaded from a server computer (not shown) via the communication unit 109 such as a wireless LAN card. The contents will be described below.

  A portable recording medium 1A (CD-ROM, MO, or the like) on which a program for causing a reader / writer (not shown) of the computer 10 shown in FIG. 29 to calculate addition data, calculate subtraction data, and add metadata is recorded. DVD-ROM or the like) is inserted, and this program is installed in the processing program 105P of the storage unit 105. Alternatively, such a program may be downloaded from an external server computer (not shown) via the communication unit 109 and installed in the storage unit 105. Such a program is loaded into the RAM 102 and executed. Thus, audio data, metadata, and specific data are input from the demultiplexer 11 via the I / F 106, and the processes described in the first to fourth embodiments are executed. The audio data to which the processed metadata and specific data are added is output to the transmission unit 18 via the I / F 108.

  The fifth embodiment has the above-described configuration, and the other configurations and operations are the same as those of the first to fourth embodiments. Therefore, the corresponding parts are denoted by the same reference numerals and detailed description thereof is omitted. To do.

  With respect to the embodiments including the first to fifth embodiments, the following additional notes are disclosed.

(Appendix 1)
In the transmission / reception device that receives the audio sound signal and transmits the received audio sound signal,
An extraction unit for extracting the first audio sound signal and the second audio sound signal according to the received audio sound signal;
A high-pass filter that extracts a frequency component of a predetermined frequency or higher from the first audio sound signal and the second audio sound signal extracted by the extraction unit;
A value related to a time-series sum signal of the first audio acoustic signal and the second audio acoustic signal related to the frequency component extracted by the high-pass filter is calculated, and addition data based on a cumulative addition value for a predetermined time of the calculated value is obtained. An adder to calculate;
A value related to a time-series difference signal between the first audio acoustic signal and the second audio acoustic signal related to the frequency component extracted by the high-pass filter is calculated, and subtraction data based on a cumulative addition value for a predetermined time of the calculated value is calculated. Subtracting part to calculate,
A first effective value calculation unit for calculating a first effective value in a time series of the first audio-acoustic signal related to the frequency component extracted by the high-pass filter;
A second effective value calculation unit for calculating a second effective value in a time series of the second audio-acoustic signal related to the frequency component extracted by the high-pass filter;
A selection unit that selects either metadata based on addition data and subtraction data, or metadata based on the first effective value and the second effective value;
When the selection unit selects metadata based on the addition data and the subtraction data, the addition unit and the subtraction data calculated by the addition unit and the subtraction unit are added to the received audio-acoustic signal as metadata, and the selection unit When metadata based on the first effective value and the second effective value is selected by the above, the first effective value and the second effective value calculated by the first effective value calculation unit and the second effective value calculation unit are metadata. An adding unit for adding to the received sound signal;
A transmission / reception apparatus comprising: a transmission unit that transmits the audio-acoustic signal to which metadata is added by the addition unit to the outside.

(Appendix 2)
In the transmission / reception device that receives the audio sound signal and transmits the received audio sound signal,
An extraction unit for extracting the first audio sound signal and the second audio sound signal according to the received audio sound signal;
A high-pass filter that extracts a frequency component of a predetermined frequency or higher from the first audio sound signal and the second audio sound signal extracted by the extraction unit;
A value related to a time-series sum signal of the first audio acoustic signal and the second audio acoustic signal related to the frequency component extracted by the high-pass filter is calculated, and addition data based on a cumulative addition value for a predetermined time of the calculated value is obtained. An adder to calculate;
A value related to a time-series difference signal between the first audio acoustic signal and the second audio acoustic signal related to the frequency component extracted by the high-pass filter is calculated, and subtraction data based on a cumulative addition value for a predetermined time of the calculated value is calculated. Subtracting part to calculate,
A first effective value calculation unit for calculating a first effective value in a time series of the first audio-acoustic signal related to the frequency component extracted by the high-pass filter;
A second effective value calculation unit for calculating a second effective value in a time series of the second audio-acoustic signal related to the frequency component extracted by the high-pass filter;
Audio that has received the addition data and subtraction data calculated by the addition unit and the subtraction unit, and the first effective value and the second effective value calculated by the first effective value calculation unit and the second effective value calculation unit as metadata. An additional unit for adding to the acoustic signal;
A transmission / reception apparatus comprising: a transmission unit that transmits the audio-acoustic signal to which metadata is added by the addition unit to the outside.

(Appendix 3)
In the transmission / reception device that receives the audio sound signal and transmits the received audio sound signal,
An extraction unit for extracting the first audio sound signal and the second audio sound signal according to the received audio sound signal;
A high-pass filter that extracts a frequency component of a predetermined frequency or higher from the first audio sound signal and the second audio sound signal extracted by the extraction unit;
A value related to a time-series sum signal of the first audio acoustic signal and the second audio acoustic signal related to the frequency component extracted by the high-pass filter is calculated, and addition data based on a cumulative addition value for a predetermined time of the calculated value is obtained. An adder to calculate;
A value related to a time-series difference signal between the first audio acoustic signal and the second audio acoustic signal related to the frequency component extracted by the high-pass filter is calculated, and subtraction data based on a cumulative addition value for a predetermined time of the calculated value is calculated. Subtracting part to calculate,
A first effective value calculation unit for calculating a first effective value in a time series of the first audio-acoustic signal related to the frequency component extracted by the high-pass filter;
A second effective value calculation unit for calculating a second effective value in a time series of the second audio-acoustic signal related to the frequency component extracted by the high-pass filter;
Metadata based on addition data and subtraction data, metadata based on first effective value and second effective value, or metadata based on addition data and subtraction data, and metadata based on first effective value and second effective value, A selection section for selecting one of
When the selection unit selects metadata based on the addition data and the subtraction data, the addition unit and the subtraction data calculated by the addition unit and the subtraction unit are added to the received audio-acoustic signal as metadata, and the selection unit When metadata based on the first effective value and the second effective value is selected by the above, the first effective value and the second effective value calculated by the first effective value calculation unit and the second effective value calculation unit are metadata. When the metadata based on the addition data and the subtraction data and the metadata based on the first effective value and the second effective value are selected by the selection unit, the addition unit and the subtraction unit And the first and second effective values calculated by the first and second effective value calculating units and the first effective value calculating unit. And adding unit for adding the speech sound signal received as,
A transmission / reception apparatus comprising: a transmission unit that transmits the audio-acoustic signal to which metadata is added by the addition unit to the outside.

It is a schematic diagram which shows the outline | summary of a transmission system. It is a block diagram which shows the hardware constitutions of a metadata calculator. It is a table which shows the value of coefficient A. It is a block diagram which shows the structure of HPF. It is a graph which shows typically a time change of left voice data and right voice data. It is a graph which shows the time change of the addition data when not passing through the process by HPF. It is a graph which shows the time change of the addition data at the time of passing through the process by HPF. It is a graph which shows the time change of the subtraction data when not passing through the process by HPF. It is a graph which shows the time change of the subtraction data at the time of passing through the process by HPF. It is a graph which shows the time change of the addition data which passed through each encoding and decoding process. It is a graph which shows the time change of the subtraction data which passed through each encoding and decoding process. It is explanatory drawing which shows the record layout of a metadata holding part. It is explanatory drawing which shows the data structure of metadata and specific data. It is a flowchart which shows the procedure of a metadata calculation process and an addition process. It is a flowchart which shows the procedure of a metadata calculation process and an addition process. It is a flowchart which shows the procedure of a metadata calculation process and an addition process. 6 is a block diagram showing a hardware configuration of a metadata calculator according to Embodiment 2. FIG. It is a graph which shows the time change of the effective value of audio | voice data. It is a graph which shows the time change of the logarithm which concerns on the effective value of audio | voice data. It is a flowchart which shows the procedure of a metadata calculation process and a conversion process. FIG. 10 is a block diagram illustrating a hardware configuration of a metadata calculator according to a third embodiment. 10 is an explanatory diagram illustrating a data structure of metadata and specific data according to Embodiment 3. FIG. 10 is a flowchart illustrating a procedure of metadata addition processing according to the third embodiment. 10 is a flowchart illustrating a procedure of metadata addition processing according to the third embodiment. It is explanatory drawing which shows the data structure of the metadata which concerns on Embodiment 4, and specific data. 14 is a flowchart illustrating a procedure of metadata addition processing according to the fourth embodiment. 14 is a flowchart illustrating a procedure of metadata addition processing according to the fourth embodiment. 14 is a flowchart illustrating a procedure of metadata addition processing according to the fourth embodiment. FIG. 10 is a block diagram illustrating a hardware configuration of a metadata calculator according to a fifth embodiment.

Explanation of symbols

DESCRIPTION OF SYMBOLS 1 Metadata calculator 1A Portable recording medium 11 Demultiplexer 12 Extraction part 13 Metadata holding part 14 Metadata calculation part 15 Metadata addition part 17 Addition part 18 Transmission part 110 HPF
112 selecting unit 141 adding unit 142 subtracting unit 143 left effective value calculating unit 144 right effective value calculating unit 151 identifier adding unit 1110 upper / lower limit converting unit 1111, 1113 integer converting unit 1112 logarithmic converting unit S1 switch S2 switch

Claims (1)

In the transmission / reception device that receives the audio sound signal and transmits the received audio sound signal,
An extraction unit for extracting the first audio sound signal and the second audio sound signal according to the received audio sound signal;
A high-pass filter that extracts a frequency component of a predetermined frequency or higher from the first audio sound signal and the second audio sound signal extracted by the extraction unit;
A value related to a time-series sum signal of the first audio acoustic signal and the second audio acoustic signal related to the frequency component extracted by the high-pass filter is calculated, and addition data based on a cumulative addition value for a predetermined time of the calculated value is obtained. An adder to calculate;
A value related to a time-series difference signal between the first audio acoustic signal and the second audio acoustic signal related to the frequency component extracted by the high-pass filter is calculated, and subtraction data based on a cumulative addition value for a predetermined time of the calculated value is calculated. Subtracting part to calculate,
A first effective value calculation unit for calculating a first effective value in a time series of the first audio-acoustic signal related to the frequency component extracted by the high-pass filter;
A second effective value calculation unit for calculating a second effective value in a time series of the second audio-acoustic signal related to the frequency component extracted by the high-pass filter;
Metadata based on addition data and subtraction data, metadata based on first effective value and second effective value, or metadata based on addition data and subtraction data, and metadata based on first effective value and second effective value, Is received by the selection unit, and when the selection unit selects metadata based on the addition data and the subtraction data, the addition data and the subtraction data calculated by the addition unit and the subtraction unit are received as metadata. When the metadata based on the first effective value and the second effective value is selected by the selection unit, the first effective value calculation unit and the second effective value calculation unit calculate the first effective value and the second effective value calculation unit. The first effective value and the second effective value are added to the audio-acoustic signal received as metadata, and the selection unit selects a method based on the addition data and the subtraction data. When data and metadata based on the first effective value and the second effective value are selected, the addition data and subtraction data calculated by the addition unit and the subtraction unit, and the first effective value calculation unit and the second effective value calculation An adding unit that adds the first effective value and the second effective value calculated by the unit to the received audio-acoustic signal as metadata;
A transmission / reception apparatus comprising: a transmission unit that transmits the audio-acoustic signal to which metadata is added by the addition unit to the outside.

Images