JP6420660B2 - Voice level monitoring apparatus and program - Google Patents

Description

  The present invention relates to a technique for measuring a loudness value for evaluating the magnitude of program audio, and in particular, an audio level monitoring apparatus and program for monitoring an audio level of program audio and calculating an auxiliary index for adjusting the audio level. About.

  An international standard method for measuring the loudness value for evaluating the size of program audio has been developed by BS ITU-R (International Telecommunication Union Radiocommunication Division). 1770-3 (see Non-Patent Document 1). Also, because the sound volume variation between programs or broadcasting stations hinders comfortable viewing by viewers, the operation regulations regarding the sound level at the time of broadcasting program production are TR by the ARIB (Radio Industry Association). -B32 (see Non-Patent Document 2).

  According to the operation regulations of ARIB TR-B32 in Non-Patent Document 2, all programs should be produced with the average loudness value of the entire program as a target loudness value, that is, -24LKFS as a target for keeping the listening level of the program appropriate. Yes, and the allowable range for operation is ± 1 dB. Here, the loudness indicates the loudness (sound sensation amount) felt by the person, and the loudness value is calculated based on the loudness measurement algorithm from the digital recording level defined in Non-Patent Document 1. Is an estimated value.

  In general, an apparatus for measuring a loudness value for evaluating the size of program audio is called a loudness meter. This loudness meter inputs an audio signal at the time of program production, measures the average value (average loudness value) of the loudness value in the entire program, and displays it on the screen as an index for evaluating the size of the program audio. Also, the average loudness value recommended in ITU-R BS.1771-1 for the past 400 milliseconds or the past 3 seconds is used as the momentary loudness value or short-term loudness value (hereinafter referred to as “instantaneous”). Some of them are displayed on the screen as “a loudness value” (see Non-Patent Document 3).

  At broadcast program production sites, program producers in charge of audio use the average loudness value up to the current time displayed on the loudness meter as an index to evaluate the size of program audio, and the average loudness value at the end of measurement. The volume of the program audio is adjusted so that the average loudness value of the entire program approaches the preset target loudness value.

Recommendation ITU-R BS. 1770-3 “Algorithms to measure audio program loudness and true-peak audio level” ARIB TR-B32 “Loudness Operation Rules for Digital Television Broadcasting Programs” Recommendation ITU-R BS. 1771-1 “Requirements for loudness and true-peak indicating meters”

  Since the average loudness value up to the current time point displayed on the loudness meter is measured using an integrated value corresponding to the elapsed time, it becomes difficult to change as time elapses. For example, when the mixing adjustment level is changed near the start of the program, and when the mixing adjustment level is changed near the end of the program (equivalent mixing adjustment level for the same audio signal as when the program is close to the start) Are compared, the average loudness value up to the present time point is likely to fluctuate in the former but hardly fluctuates in the latter.

  For this reason, as the current time approaches the end of the measurement at the end of the program, the program producer becomes difficult to reflect the adjustment value of the audio level in the fluctuation of the average loudness value and the average loudness value at the end of the measurement. There has been a problem that it is difficult to adjust the sound level so that the loudness value approaches the target loudness value.

  Accordingly, the present invention has been made to solve the above-described problems, and provides an audio level monitoring apparatus and program capable of calculating an auxiliary index useful for adjusting the audio level of a program.

  In order to solve the above-mentioned problems, an audio level monitoring apparatus according to the present invention is an audio level monitoring apparatus that monitors an audio level of a program, inputs an audio signal of the program, and starts from a predetermined measurement start time to a current time. Average loudness value calculating means for calculating the average loudness value of the current time as the average loudness value at the current time point, the average loudness value at the current time point calculated by the average loudness value calculating means, and the average loudness value of the entire program And the average loudness in the remaining time from the current time point to the measurement end time point based on the target loudness value indicating the target value and the measurement time length from the measurement start time point to the predetermined measurement end time point. The average loudness value at the end of the measurement calculated by the value calculating means is brought close to the target loudness value. The data, characterized by comprising a an auxiliary index calculating means for calculating as an auxiliary index for adjusting the sound level.

  In the audio level monitoring apparatus according to the present invention, the auxiliary index calculating means may be a target loudness indicating a target value of the average loudness value at the current time point calculated by the average loudness value calculating means and the average loudness value of the entire program. The remaining time target average that can be the target of the average loudness value in the remaining time from the current time to the measurement end time based on the value and the measurement time length from the measurement start time to the predetermined measurement end time A loudness value is calculated as the auxiliary index.

  In the audio level monitoring apparatus according to the present invention, the auxiliary index calculating means may be an allowable upper limit indicating an allowable upper limit of the average loudness value at the current time point calculated by the average loudness value calculating means and the average loudness value of the entire program. Based on the loudness value and the measurement time length from the measurement start time to the predetermined measurement end time, the remaining time that can be an allowable upper limit of the average loudness value in the remaining time from the current time to the measurement end time An allowable upper limit value of the target average loudness value is calculated as the auxiliary index.

  In the audio level monitoring apparatus according to the present invention, the auxiliary index calculating means indicates an allowable lower limit indicating an allowable lower limit of the average loudness value at the current time point calculated by the average loudness value calculating means and the average loudness value of the entire program. Based on the loudness value and the measurement time length from the measurement start time to the predetermined measurement end time, the remaining time that can be an allowable lower limit of the average loudness value in the remaining time from the current time to the measurement end time An allowable lower limit value of the target average loudness value is calculated as the auxiliary index.

  Further, in the audio level monitoring apparatus according to the present invention, the auxiliary index calculation means may calculate the target loudness value and the average loudness value of the entire program in the measurement time length from the measurement start time to the measurement end time. An excess allowable limit amount indicating a difference between an allowable upper limit loudness value indicating an allowable upper limit and an average loudness value calculated by the average loudness value calculation means in the time length from the measurement start time to the current time point is the target. A ratio between the average loudness value when the loudness value is equal to or greater than the loudness value and an excess amount indicating a difference between the target loudness values is calculated as an excess loudness value, and the target loudness value in the measurement time length, and the The difference in the lower limit loudness value indicating the lower limit of the average loudness value of the entire program And the average loudness when the average loudness value calculated by the average loudness value calculation means in the time length from the measurement start time to the current time is less than or equal to the target loudness value. A ratio between a value and an insufficient amount indicating a difference between the target loudness values is calculated as an insufficient loudness value, and the excess loudness value and the insufficient loudness value are used as the auxiliary index.

  Also, the audio level monitoring apparatus according to the present invention has one side that indicates that the excess loudness value and the insufficient loudness value calculated by the auxiliary index calculation means are exceeded and a side that indicates a shortage. When the average loudness value at the current time point coincides with the target loudness value, the excess / underdue loudness value display means for displaying with the level meter is displayed. When the average loudness value at the current time point is larger than the target loudness value and less than or equal to the allowable upper limit loudness value, the level is displayed in the first form on the side indicating the excess in the level meter. , The average loudness value at the current time point is greater than the target loudness value, and the allowable upper limit label If it is larger than the loudness value, a level is displayed in a form different from the first on the side indicating the excess in the level meter, and the average loudness value at the current time point is smaller than the target loudness value. And when it is equal to or greater than the allowable lower limit loudness value, a level is displayed in the second form on the side indicating the lack in the level meter, and the average loudness value at the current time point is smaller than the target loudness value. And when it is smaller than the allowable lower limit loudness value, a level is displayed in a form different from the second on the side indicating the lack in the level meter.

  Further, the sound level monitoring device according to the present invention, at the measurement start time, the measurement time length from the measurement start time to the measurement end time is a fixed time length, and when a switching signal is input during measurement, The measurement end time is set to a time after a predetermined time from the current time, and the measurement time length from the measurement start time to the measurement end time is a variable time length. .

  Furthermore, the program according to the present invention causes a computer to function as the sound level monitoring apparatus.

  As described above, according to the present invention, an auxiliary index useful for adjusting the audio level of a program can be calculated. For example, when adjusting the audio level of a program, a program producer can easily approximate the average loudness value of the entire program to a preset target loudness value by referring to the calculated auxiliary index. .

It is a figure which shows the transition image of the square average value for every block at the time of calculating remaining time target average loudness value. It is a figure which shows the transition image of the root mean square value for every block at the time of calculating an excess loudness value and an insufficient loudness value. It is a block diagram which shows the structure of the audio | voice level monitoring apparatus by embodiment of this invention. It is a flowchart which shows an example of the root mean square value calculation process for every block by an average loudness value calculation means. It is a flowchart which shows the absolute gating process by an average loudness value calculation means. It is a flowchart which shows the relative gating process by an average loudness value calculation means. It is a block diagram which shows the structure of a remaining time target average loudness value calculation means. It is a flowchart which shows the process of a remaining time target average loudness value calculation means. It is a block diagram which shows the structure of an excess / insufficient loudness value calculation means. It is a flowchart which shows the process of an excess / insufficient loudness value calculation means. It is a figure which shows an example of the display form of the excess loudness value displayed on the screen by the excess / insufficient loudness value display means and the insufficient loudness value.

  Hereinafter, embodiments for carrying out the present invention will be described in detail with reference to the drawings. In the present invention, an auxiliary loudness index (hereinafter referred to as an auxiliary index) for bringing the average loudness value of the entire program close to a preset target loudness value is referred to as an average loudness value at the current time point and a measurement start time point. The calculation is based on the measurement time length until the end of measurement. For example, the present invention provides, as auxiliary indicators, a remaining time target average loudness value, an allowable upper limit value of the remaining time target average loudness value, an allowable lower limit value of the remaining time target average loudness value, and an excess loudness value and an insufficient loudness value. Calculate at least one of These auxiliary indicators are used for program production and the like.

[Remaining time target average loudness value]
First, a method for calculating a remaining time target average loudness value as an auxiliary index will be described. When using a conventional loudness meter, a program producer uses an average loudness value (an average loudness value up to the current time point, hereinafter the average loudness value from the measurement start time (program start time) to the current time). “The average loudness value at the current time point”) is used as an index, and the program at the current time point is such that the average loudness value (the average loudness value of the entire program) at the end of measurement (the program end time) approaches the target loudness value. Adjust the sound level.

  However, the adjustment of the average loudness value after the current time depends on the remaining time until the end of the program, and the degree to which the instantaneous loudness value affects the average loudness value of the entire program is different. Even if the sound level is adjusted using only the index as an index, it is difficult to bring the average loudness value of the entire program close to the target loudness value.

  Therefore, in the present invention, it is an auxiliary index for compensating for the excess or deficiency of the average loudness value at the current time point relative to the target loudness value, and becomes a target at the remaining time from the current time point to the end point of measurement. The obtained average loudness value is calculated as the remaining time target average loudness value. As a result, the program producer can use the remaining target time average loudness value as an auxiliary index for bringing the average loudness value of the entire program at the end of the measurement closer to the target loudness value. The adjustment becomes easier.

  FIG. 1 is a diagram showing a transition image of the mean square value for each block when calculating the remaining time target average loudness value. The horizontal axis indicates the elapsed time (the number of blocks from which the input speech has been cut out), and the vertical axis indicates the mean square value in the block of the input speech that has been subjected to the K characteristic filter processing.

とする。Ｎ_allは正の整数であり、ｋ，ｍは０以上の整数であり、ｋ≧ｍである。現在時点における番組全体の見込み有効ブロック数Ｎは、以下の式で表される。
［数１］
Ｎ＝Ｎ_all−ｋ＋ｍ ・・・（１） The measurement time length indicating the number of blocks of the entire program is N _all , the number of blocks (block number) at the present time is k, the number of effective blocks after the absolute gating process and the relative gating process up to the current time is m, and the remaining time Target square mean value

And N _all is a positive integer, k and m are integers of 0 or more, and k ≧ m. The expected number of effective blocks N of the entire program at the current time is expressed by the following equation.
[Equation 1]
N = N _all −k + m (1)

、ターゲットラウドネス値をＬ_t（＝−２４ＬＫＦＳ）、ターゲットラウドネス値Ｌ_tに対応する二乗平均値（ターゲット二乗平均値）をＺ_tとする。現在時点から測定終了時点までの間の残時間において、番組全体の二乗平均値（測定終了時点の二乗平均値）に対応する番組全体の平均ラウドネス値をターゲットラウドネス値Ｌ_tに一致させるために、目標とすべき二乗平均値（残時間目標二乗平均値）は、以下の式で表される。
［数２］

Also, the mean square value of the audio signal from the measurement start time to the current time (the mean square value at the current time)

The target loudness value is L _t (= −24LKFS), and the mean square value (target mean square value) corresponding to the target loudness value L _t is Z _t . In order to make the average loudness value of the entire program corresponding to the mean square value of the entire program (the mean square value of the measurement end point) corresponding to the target loudness value L _t in the remaining time from the current time point to the measurement end time point, The mean square value to be targeted (remaining time target mean square value) is expressed by the following equation.
[Equation 2]

この式（３）は、二乗平均値とラウドネス値との間の関係を定義した数式である。 By converting this into loudness, a remaining time target average loudness value L ₀ is calculated. That is, in order to make the average loudness value of the entire program corresponding to the mean square value of the entire program (the mean square value at the end of measurement) match the target loudness value L _t during the remaining time, the target loudness value (remaining value) Time target average loudness value: a value that can be a target in the remaining time) L ₀ is expressed by the following equation.
[Equation 3]

This equation (3) is an equation that defines the relationship between the mean square value and the loudness value.

As described above, according to the operation rules of ARIB TR-B32 of Non-Patent Document 2, the average loudness value of the entire program is an allowable value of ± 1 LKFS (allowable upper limit loudness) with respect to the target loudness value L _t (= −24LKFS). A value L _u (= −23LKFS) and an allowable lower limit loudness value L _l (= −25LKFS)) are defined. The allowable upper limit loudness value L _u indicates the value of the allowable upper limit of the average loudness value of the entire program, and the allowable lower limit loudness value L _l indicates the value of the allowable lower limit of the average loudness value of the entire program. Thus, in addition to using the remaining time target average loudness value L ₀ as an auxiliary indicator, the allowable upper limit value and / or the allowable lower limit value of the remaining time target average loudness value L ₀ may be used as an auxiliary indicator. Further, instead of the remaining time target average loudness value L ₀ and the auxiliary indicator, the allowable upper limit value of the remaining time target average loudness value L ₀ and / or the allowable lower limit value may be an auxiliary index. Allowable upper limit of remaining time target average loudness value L ₀ is a permissible upper limit and becomes possible values of the average loudness value in the remaining time, the allowable lower limit value of the remaining time target average loudness value L ₀ is the average loudness value in remaining time It is a value that can be an allowable lower limit.

Similar to the remaining time target average loudness value L ₀ , the allowable upper limit value and the allowable lower limit value of the remaining time target average loudness value L ₀ are calculated using the equations (2) and (3). Allowable upper limit of remaining time target mean square value, by the formula (2), mean square values corresponding to the target loudness value L _t (target mean square value) instead of Z _t, corresponding to the allowable upper limit loudness value L _u The allowable upper limit value of the remaining time target average loudness value L ₀ is calculated by the above equation (3) in which the relationship between the mean square value and the average loudness value is defined. It is calculated by converting the allowable upper limit value of the mean square value. Further, the allowable lower limit value of the remaining time target mean square value is obtained by the above-described equation (2), instead of the target mean square value Z _t corresponding to the target loudness value L _t , the mean square value corresponding to the permissible lower limit loudness value L _l. The allowable lower limit value of the remaining time target average loudness value L ₀ is calculated by converting the allowable lower limit value of the remaining time target mean square value according to the equation (3).

Accordingly, since it is possible to have a tolerance to the auxiliary indicator, program producer utilizes average loudness value of the entire program of the measuring end as auxiliary indicator having a tolerance for approximating the target loudness value L _t This makes it easier to adjust the audio level in consideration of the entire program.

[Excessive loudness value / Insufficient loudness value]
Next, a method for calculating an excess loudness value and an insufficient loudness value as an auxiliary index will be described. As described above, according to the operation regulations of ARIB TR-B32 in Non-Patent Document 2, the allowable upper limit value and the allowable lower limit value (allowable upper limit loudness value) of the average loudness value of the entire program with respect to the target loudness value L _{t are} ± 1 LKFS. The value L _u and the allowable lower limit loudness value L _l ) are defined. The product of the allowable upper limit loudness value L _u and the program time length is the average loudness value L _k at the current time point (the average loudness value L _{k with} respect to the target loudness value L _t from the measurement start time to the current time point). It can be understood that this is the upper limit of the total sum. The value of the product of the time length of the allowable lower loudness value L _l and the program may be regarded as the lower limit of the permissible value of the sum of the average loudness value L _k of the current time.

Therefore, in the present invention, at the current time point, the sum of excess and deficiency of the average loudness value L _{k with} respect to the target loudness value L _t is calculated, respectively, and the allowable upper limit loudness value L _u and allowable lower limit loudness value L _l. And the time length of the program are calculated, and the ratio (excessive loudness value ELL and insufficient loudness value SLL) is calculated as an auxiliary index.

That is, in the present invention, in the measurement time length, and the allowable limit amount of excess indicating a difference of the target loudness values L _t and the allowable upper limit loudness value L _u, the current from the start of measurement time points until the measurement end time from the start of measurement in the time length until the average loudness value L _k is the ratio between the excess amount indicating the difference of the average loudness value L _k and target loudness value L _t in the case where not less than the target loudness value L _t, excess loudness Calculated as value ELL. Further, in the present invention, an insufficient allowable amount indicating a difference between the target loudness value L _t and the allowable lower limit loudness value L _l in the measurement time length from the measurement start time to the measurement end time, and the current time from the measurement start time in the time length until the ratio between the deficient amount indicating the difference between the average loudness value L _k and target loudness value L _t where the average loudness value L _k is equal to or less than the target loudness value L _t, lack loudness Calculated as value SLL.

As a result, the program producer can know the degree of excess and deficiency of the average loudness value L _k at the current time point with respect to the loudness value allowable limit of the entire program. Accordingly, program producer can utilize the average loudness value of the entire program of the measuring end as an auxiliary indication to approximate the target loudness value L _t, it is easy to adjust the sound level in consideration of the entire program.

FIG. 2 is a diagram showing a transition image of the mean square value for each block when calculating the excess loudness value ELL and the insufficient loudness value SLL. The horizontal axis, the vertical axis, the measurement time length N _all indicating the number of blocks in the entire program, the number k of blocks at the current time point, and the number m of effective blocks are the same as those in FIG. The target mean square value corresponding to the target loudness value L _t is Z _t , the mean square value corresponding to the allowable upper limit loudness value L _u (allowable upper limit mean square value) is Z _u , and the mean square corresponding to the allowable lower limit loudness value L _l The value (allowable lower limit mean square value) is Z _l , and the current mean square mean value (measured mean square value) is Z _i .

The sum of the excess of the mean square value Z _{i based on} the target mean square value Z _t and the sum of all the blocks from the measurement start time to the current block number i is the maximum allowable excess The ratio between them (excess allowable upper limit ratio) is expressed by the following equation.
[Equation 4]

Further, the sum of the shortage of the mean square value Z _{i based on} the target mean square value Z _t added in each block from the measurement start time to the current block number i, and the maximum allowable shortage The ratio between the two (deficiency allowable lower limit ratio) is expressed by the following equation.
[Equation 5]

The equation (4) can be rewritten as follows.
[Equation 6]

The equation (5) can be rewritten as follows.
[Equation 7]

The expression (6) is expressed on a logarithmic scale, and this is called an excess loudness value ELL (Excess Loudness Level).
[Equation 8]

Further, the expression (7) is expressed on a logarithmic scale, and this is referred to as an insufficient loudness value SLL (Shortage Loudness Level).
[Equation 9]

The excess loudness value ELL and the insufficient loudness value SLL are values that vary with the same degree without depending on the elapsed time with respect to the audio level adjusted by the program producer, and are linear except for the influence of the gating process. Fluctuates. On the other hand, when the average loudness value L _k at the current time point is calculated at a time point close to the start of measurement and when calculated at a time point close to the end of the measurement, the degree of variation with respect to the sound level is different. That is, when the average loudness value L _k at time close to the measurement start is calculated, along with the variation of the sound level will vary in conjunction with the average loudness value L _k, average loudness value L _k at time close to end of measurement Is calculated, the average loudness value L _k does not change much even if the sound level changes. That is, the average loudness value L _k at the present time is a value that varies with a different degree depending on the elapsed time with respect to the audio level adjusted by the program producer, and does not vary linearly.

Therefore, the program producer can easily adjust the audio level by referring to the excess loudness value ELL and the insufficient loudness value SLL as auxiliary indicators as compared with the case of referring to the average loudness value L _k at the current time point as an indicator. .

Thus, in the present invention, at the time of program production, remaining time target average loudness value L _0, the allowable upper limit value of the remaining time target average loudness value L _0, the remaining time target average loudness value permissible lower limit value of L _0, and, At least one of the excess loudness value ELL and the insufficient loudness value SLL is calculated as an auxiliary indicator. Thus, the program producer can easily approximate the measured average loudness value of the entire program to the preset target loudness value L _t by referring to the auxiliary index. That is, the program producer can adjust the audio level in consideration of the entire program time length, which cannot be obtained only by the average loudness value L _k up to the present time point.

[Voice level monitoring device]
Next, an audio level monitoring apparatus according to an embodiment of the present invention will be described. FIG. 3 is a block diagram showing the configuration of the audio level monitoring apparatus. The audio level monitoring apparatus 1 is configured so that a user such as a program producer sets a program time when a measurement time length N _all that is a time length to be measured and a target loudness value L _t that is a target loudness value are set in advance. This is a device that monitors the audio level of the program, calculates an auxiliary index for adjusting the audio level of the program, and displays it on the screen.

  The sound level monitoring apparatus 1 includes an average loudness value calculating means 10, an average loudness value displaying means 11, a prospective effective block number calculating means 12, a remaining time target average loudness value calculating means 13, a remaining time target average loudness value displaying means 14, and an excess. A deficient loudness value calculating unit 15 and an excess / deficient loudness value display unit 16 are provided. The remaining time target average loudness value calculating means 13 and the excess / insufficient loudness value calculating means 15 constitute auxiliary index calculating means.

The audio level monitoring apparatus 1 receives an input audio signal (audio signal of a predetermined number of channels), a target loudness value L _t , a measurement time length N _all , an allowable upper limit loudness value _Lu, and an allowable lower limit loudness value L _l. The Then, the average loudness value L _k at the current time point is calculated by the voice level monitoring device 1, and the remaining time target average loudness value L ₀ , the allowable upper limit value of the remaining time target average loudness value L ₀ , and the remaining time target are used as auxiliary indicators. The allowable lower limit value of the average loudness value L ₀ and at least one of the excess loudness value ELL and the insufficient loudness value SLL are calculated and displayed on the screen.

The target loudness value L _t , the measurement time length N _all , the allowable upper limit loudness value L _u and the allowable lower limit loudness value L _l are input to the sound level monitoring apparatus 1 as preset values.

[Average loudness value calculation means 10]
The average loudness value calculation means 10 receives an input audio signal, and the recommended ITU-R BS. In accordance with the provisions of 1770-3, to calculate the average loudness value L _k between until the present time (average loudness value L _k of the current time) from the start of measurement, and calculates the number of valid blocks m to be described later.

Specifically, the average loudness value calculating means 10 calculates a mean square value Z _k of the audio signal for each block in units of blocks having a predetermined time length (performs a mean square value calculation process for each block).

Next, the average loudness value calculation means 10 uses the mean square value Z _k calculated for each block, and discards the mean square value Z _k of the block that can be regarded as silence of a predetermined magnitude or less by threshold processing (absolute gain). Process).

、現在時点の平均ラウドネス値Ｌ_k及び有効ブロック数ｍを算出する（相対ゲーティング処理を行う）。 Next, the average loudness value calculation means 10 uses the mean square value Z _i of the blocks that have not been discarded in the absolute gating process, and the mean square value of the blocks that can be regarded as a small sound having a predetermined volume or less by the threshold process. Z _i is discarded and the mean square value Z _i of the blocks that have not been discarded is used to calculate the mean square value between the measurement start time and the current time (the mean square value at the current time).

The average loudness value L _k and the number m of effective blocks at the current time are calculated (relative gating processing is performed).

は、残時間目標平均ラウドネス値算出手段１３及び超過・不足ラウドネス値算出手段１５に出力される。また、平均ラウドネス値算出手段１０により算出された現在時点の平均ラウドネス値Ｌ_kは、平均ラウドネス値表示手段１１に出力される。 The effective block number m calculated by the average loudness value calculating unit 10 is output to the expected effective block number calculating unit 12, the remaining time target average loudness value calculating unit 13, and the excess / deficient loudness value calculating unit 15. Further, the mean square value corresponding to the current average loudness value L _k calculated by the average loudness value calculating means 10.

Is output to the remaining time target average loudness value calculation means 13 and the excess / deficiency loudness value calculation means 15. The average loudness value L _k calculated by the average loudness value calculating unit 10 is output to the average loudness value display unit 11.

  FIG. 4 is a flowchart showing an example of a mean square value calculation process for each block by the average loudness value calculation means 10, and shows an example of input sound of five channels. The average loudness value calculation means 10 inputs the audio signal of the input audio of 5 channels (step S401), and cuts out the audio signal of the block with a time length of 400 ms as a block for each channel (step S402).

The average loudness value calculation means 10 performs a K characteristic filter process on the audio signal of the block for each channel (step S403), calculates a mean square value (step S404), and is set in advance for the mean square value. Multiply the weighting coefficient for each channel (step S405), and add the multiplication results for each channel in all channels (step S406). Then, the average loudness value calculation means 10 increments the number k of blocks corresponding to the current time (step S407), and generates the addition result of step S406 as the mean square value Z _k of the blocks at the current time (step S408).

  In addition, the recommendation ITU-R BS. According to the provisions of 1770-3, the block of input speech is shifted every 100 ms. Therefore, in the mean square value calculation process for each block, the process is performed with the audio signals of the overlapping blocks before and after the current time point. The number of blocks k is set to 0 at the measurement start time, and thereafter counts up sequentially.

In this way, the average loudness value calculation means 10 performs the mean square value of the audio signal (the mean square value of the kth block) for each block in units of blocks of a predetermined time length by the mean square value calculation process for each block. ) Calculate Z _k . Then, using the root mean square value Z _k of the k-th block, an absolute gating process shown in FIG. 5 described later and a relative gating process shown in FIG. 6 described later are performed.

FIG. 5 is a flowchart showing the absolute gating process by the average loudness value calculation means 10. The average loudness value calculation means 10 calculates the mean square value Z _k of the kth block calculated by the mean square value calculation process for each block by the following formula, and the loudness value (momentary loudness value) of the kth block. Conversion to LL _k (steps S501 and S502).
[Equation 10]
LL _k = −0.691 + ₁₀ log ₁₀ Z _k (10)
The equation (10) is an equation that defines the relationship between the root mean square value Z _k and the loudness value LL _k, and is the same as the equation (3).

The average loudness value calculation means 10 determines whether or not the loudness value LL _k of the k-th block is larger than −70LKFS (step S503), and determines that the loudness value LL _k is not larger than −70LKFS (step S503). Step S503: N), the root mean square value Z _k of the k-th block is discarded (Step S504). As a result, the audio signal of the k-th block is regarded as silence of −70 LKFS or less, and the mean square value Z _k is discarded because it does not contribute to the sound volume of the entire program for audibility.

On the other hand, when the average loudness value calculation means 10 determines that the loudness value LL _k is larger than −70LKFS (step S503: Y), the mean square value Z _k of the k-th block is shown in FIG. The number of valid blocks n is incremented (step S506). As a result, the audio signal of the kth block is considered not to be silent, and its mean square value Z _k is stored in the memory, as it contributes to the loudness of the entire program for audibility. The effective block number n is set to 0 at the start of the process, and is counted up in step S506.

を算出し（ステップＳ５０７）、図３には図示していないメモリに保存する（ステップＳ５０８）。 The average loudness value calculation means 10 calculates the mean square value of the number n of effective blocks in the kth block.

Is calculated (step S507) and stored in a memory not shown in FIG. 3 (step S508).

を読み出し、これにｋ番目のブロックの二乗平均値Ｚ_kを加算し、加算結果を有効ブロック数ｎで除算し、除算結果をｋ番目のブロックにおける有効ブロック数ｎの二乗平均値

として求める。 Specifically, the average loudness value calculation means 10 calculates the mean square value of the number of effective blocks (n−1) stored in step S508 from a memory not shown in FIG.

, The mean square value Z _k of the _kth block is added to this, the addition result is divided by the number of valid blocks n, and the result of division is the mean square value of the number of valid blocks n in the kth block

Asking.

Thus, the average loudness value calculating means 10 uses the mean square value Z _k calculated for each block by the absolute gating process, and the mean square which can be regarded as silence of a predetermined magnitude or less in the threshold process. Discard the value Z _k .

を、ラウドネス値Ｌ_aに変換する（ステップＳ６０１，ステップＳ６０２）。 FIG. 6 is a flowchart showing the relative gating process by the average loudness value calculation means 10. The average loudness value calculating means 10 uses the equation (10) to calculate the mean square value of the number n of effective blocks in the kth block calculated by the absolute gating process.

Is converted into _a loudness value La (step S601, step S602).

  The average loudness value calculation means 10 sets 0 to the index i and sets 0 to the number m of effective blocks (step S603). Then, the average loudness value calculation means 10 determines whether or not the index i is smaller than the block number k (step S604), and when it is determined that the index i is smaller than the block number k (step S604: Y), The process proceeds to step S606.

The average loudness value calculation means 10 reads the mean square value Z _i of the i-th block stored by the process of step S505 in FIG. 5 from the memory, and uses the above equation (10) to calculate the mean square of the i-th block. The value Z _i is converted into the loudness value LL _i (step S605).

Average loudness value calculating unit 10 shifts from step S604 and step S605, subtracts the 10LKFS from loudness value L _a, loudness value LL _i of i-th block is determined whether greater than the subtraction result When it is determined that the loudness value LL _i is not larger than the subtraction result (step S606: N) (step S606), the root mean square value Z _i of the i-th block is discarded (step S607). As a result, the sound signal of the i-th block is regarded as a small sound, and its mean square value Z _i is discarded.

On the other hand, if the average loudness value calculation means 10 determines that the loudness value LL _i is larger than the subtraction result (step S606: Y), the average loudness value m is incremented (step S608). As a result, the audio signal of the i-th block is regarded as not being silent and not being a small sound.

を算出し（ステップＳ６０９）、図３には図示していないメモリに保存する（ステップＳ６１０）。 The average loudness value calculation means 10 calculates the mean square value of the number m of effective blocks in the i-th block.

Is calculated (step S609) and stored in a memory not shown in FIG. 3 (step S610).

を読み出し、読み出した二乗平均値にｉ番目のブロックの二乗平均値Ｚ_iを加算し、加算結果を有効ブロック数ｍで除算し、除算結果をｉ番目のブロックにおける有効ブロック数ｍの二乗平均値

として求める。 Specifically, the average loudness value calculation means 10 calculates the mean square value of the number of effective blocks (m−1) stored in step S610 from a memory not shown in FIG.

, Add the mean square value Z _i of the i th block to the read mean square value, divide the addition result by the number m of effective blocks, and divide the result of the square mean value of the number m of effective blocks in the i th block

Asking.

  The average loudness value calculation means 10 increments the index i (step S611), and proceeds to step S604.

を読み出し、これを

とし、前記式（１０）を用いて、これを現在時点の平均ラウドネス値Ｌ_kに変換する（ステップＳ６１２）。 On the other hand, when the average loudness value calculation means 10 determines in step S604 that the index i is not smaller than the number k of blocks (step S604: N), the average loudness value calculation means 10 in the kth block from the memory not shown in FIG. Mean square of m effective blocks

Read this

And using the equation (10), this is converted into the average loudness value L _k at the present time (step S612).

、現在時点の平均ラウドネス値Ｌ_k及び有効ブロック数ｍを算出する。 In this way, the average loudness value calculation means 10 can be regarded as a small sound having a predetermined magnitude or less by the threshold process using the mean square value Z _i that has not been discarded in the absolute gating process by the relative gating process. The mean square value Z _i is discarded and the mean square value Z _i that has not been discarded is used to calculate the mean square value at the present time.

Then, the average loudness value L _k and the effective block number m at the current time point are calculated.

Returning to FIG. 3, the average loudness value display unit 11, from the average loudness value calculating unit 10 inputs the average loudness value L _k of the current time, the screen displays the average loudness value L _k of the current time.

[Expected effective block number calculation means 12]
The expected effective block number calculating means 12 inputs a preset measurement time length N _all and also inputs the effective block number m from the average loudness value calculating means 10, and the entire program at the current time point according to the equation (1). The expected effective number of blocks N is calculated. The estimated effective block number N includes the effective block number m when it is assumed that all blocks in the remaining time are effective by the absolute gating process shown in FIG. 5 and the relative gating process shown in FIG. Indicates the number of blocks. The expected effective block number N calculated by the expected effective block number calculating means 12 is output to the remaining time target average loudness value calculating means 13 and the excess / deficient loudness value calculating means 15.

及び有効ブロック数ｍを、見込み有効ブロック数算出手段１２から見込み有効ブロック数Ｎをそれぞれ入力する。そして、残時間目標平均ラウドネス値算出手段１３は、これらの値に基づいて、補助指標である残時間目標平均ラウドネス値Ｌ₀を算出する。残時間目標平均ラウドネス値算出手段１３により算出された残時間目標平均ラウドネス値Ｌ₀は、残時間目標平均ラウドネス値表示手段１４に出力される。 [Remaining time target average loudness value calculation means 13]
The remaining time target average loudness value calculating means 13 inputs a preset target loudness value L _t and the mean square value of the current time point from the average loudness value calculating means 10.

And the number m of effective blocks are input as the number N of expected effective blocks from the expected effective block number calculation means 12. Then, the remaining time target average loudness value calculating means 13 calculates a remaining time target average loudness value L ₀ as an auxiliary index based on these values. The remaining time target average loudness value L ₀ calculated by the remaining time target average loudness value calculating means 13 is output to the remaining time target average loudness value display means 14.

Incidentally, the remaining time target average loudness value calculating means 13, not shown in FIG. 3, the allowable upper limit loudness values enter the L _u and allowable lower loudness value L _l, the remaining time target average loudness for allowable upper loudness value L _u An allowable upper limit value of value L ₀ and an allowable lower limit value of remaining time target average loudness value L ₀ with respect to allowable lower limit loudness value L ₁ are also calculated. The allowable upper limit value and the allowable lower limit value of the remaining time target average loudness value L ₀ calculated by the remaining time target average loudness value calculating means 13 are output to the remaining time target average loudness value display means 14.

  FIG. 7 is a block diagram showing the configuration of the remaining time target average loudness value calculating means 13, and FIG. 8 is a flowchart showing the processing of the remaining time target average loudness value calculating means 13. The remaining time target average loudness value calculating means 13 includes a loudness value / square average value conversion stage 20, a memory 21, a remaining time target square average value calculation stage 22, and a square average value / loudness value conversion stage 23.

The loudness value / square mean value conversion stage 20 inputs a preset target loudness value L _t (step S801), and uses the equation (10) to convert the target loudness value L _t to the target mean square value Z _t . The data is converted (step S802) and stored in the memory 21.

The loudness value / square mean value conversion stage 20 converts the allowable upper limit loudness value L _u and the allowable lower limit loudness value L _{1 into the} allowable upper limit mean square value and the allowable lower limit mean square value, respectively, using the equation (10). And stored in the memory 21.

及び有効ブロック数ｍを、見込み有効ブロック数算出手段１２から見込み有効ブロック数Ｎを入力し（ステップＳ８０３）、これらの値をメモリ２１に保存する。 The remaining time target average loudness value calculation means 13 receives the mean square value at the present time from the average loudness value calculation means 10.

Further, the effective block number m is input as the expected effective block number N from the expected effective block number calculating means 12 (step S803), and these values are stored in the memory 21.

、有効ブロック数ｍ及び見込み有効ブロック数Ｎを読み出し、前記式（２）により、残時間目標二乗平均値

を算出する（ステップＳ８０４）。残時間目標二乗平均値算出段２２により算出された残時間目標二乗平均値

は、二乗平均値／ラウドネス値変換段２３に出力される。 The remaining time target mean square value calculation stage 22 reads the target mean square value Z _t from the memory 21 and the mean square value at the current time point.

The effective block number m and the expected effective block number N are read out, and the remaining time target mean square value is calculated by the above equation (2)

Is calculated (step S804). The remaining time target mean square value calculated by the remaining time target mean square value calculation stage 22

Is output to the root mean square value / loudness value conversion stage 23.

  The remaining time target mean square value calculation stage 22 also reads the allowable upper limit mean square value and the allowable lower limit mean square value from the memory 21, and the allowable upper limit value and the allowable lower limit value of the remaining time target mean square value according to the above equation (2). Calculate the value. The allowable upper limit value and the allowable lower limit value of the remaining time target mean square value calculated by the remaining time target mean square value calculation stage 22 are also output to the mean square value / loudness value conversion stage 23.

を入力し、前記式（１０）を用いて、これを残時間目標平均ラウドネス値Ｌ₀に変換する（ステップＳ８０５）。二乗平均値／ラウドネス値変換段２３により変換された残時間目標平均ラウドネス値Ｌ₀は、残時間目標平均ラウドネス値表示手段１４に出力される（ステップＳ８０６）。 The root mean square value / loudness value conversion stage 23 receives the remaining time target square mean value from the remaining time target square mean value calculation stage 22.

Is converted into the remaining time target average loudness value L ₀ using the above-described equation (10) (step S805). The remaining time target average loudness value L ₀ converted by the mean square value / loudness value conversion stage 23 is output to the remaining time target average loudness value display means 14 (step S806).

The mean square value / loudness value conversion stage 23 also inputs the allowable upper limit value and the allowable lower limit value of the remaining time target mean square value from the remaining time target mean square value calculation stage 22, and uses the above equation (10), These are converted into an allowable upper limit value and an allowable lower limit value of the remaining time target average loudness value L ₀ . The allowable upper limit value and the allowable lower limit value of the remaining time target average loudness value L ₀ converted by the mean square value / loudness value conversion stage 23 are also output to the remaining time target average loudness value display means 14.

Returning to FIG. 3, the remaining time target average loudness value display means 14 sends the remaining time target average loudness value L ₀ , the allowable upper limit value of the remaining time target average loudness value L ₀ , and the remaining time target average loudness value L _0. Input allowable lower limit values and display them on the screen as auxiliary indicators.

Thus, the remaining time target average loudness value calculating means 13 calculates the remaining time target average loudness value L _0, which is the loudness value to be targeted at the remaining time, at the current time point, and the remaining time target average loudness value displaying means. 14 is displayed on the screen as an auxiliary index. This remaining time target average loudness value L ₀ targets the average loudness value of the entire program at the end of measurement by matching the average loudness value from the current time to the end of measurement with the remaining time target average loudness value L _0. It is a value that makes it possible to approach the loudness value L _t . However, in practice, the average loudness value from the current time point to the measurement end time point is a future value after the current time point, and therefore can only be predicted from instantaneous fluctuations in the loudness value.

Therefore, the program producer refers to the remaining time target average loudness value L ₀ displayed on the screen, and makes the average value of the instantaneous loudness value coincide with the remaining time target average loudness value L ₀ after the current time point. by adjusting the sound level, it can be easily brought close to the average loudness value of the whole program in the measurement end to the target loudness value L _t.

Further, the remaining time target average loudness value calculating means 13 allows the remaining time target average loudness value L ₀ when the allowable upper limit loudness value L _u and the allowable lower limit loudness value L _l are targeted at the remaining time at the present time. The upper limit value and the allowable lower limit value are calculated and displayed on the screen as auxiliary indicators by the remaining time target average loudness value display means 14. The allowable upper limit value of the remaining time target average loudness value L ₀ is set so that the average loudness value from the current time point to the measurement end time point is less than or equal to the allowable upper limit value of the remaining time target average loudness value L ₀ . is a value that makes it possible to the average loudness value of the entire program below the allowable upper limit loudness value L _u. Further, the allowable lower limit value of the remaining time target average loudness value L ₀ is the average loudness value from the current time to the measurement end time by more than the allowable lower limit value of the remaining time target average loudness value L _0, the measurement end is a value that makes it possible to the average loudness value of the entire program than allowable lower loudness value L _l at the time. However, in practice, the average loudness value from the current time point to the measurement end time point is a future value after the current time point, and therefore can only be predicted from instantaneous fluctuations in the loudness value.

Therefore, the program creator refers to the allowable upper limit value and the allowable lower limit value of the remaining time target average loudness value L ₀ displayed on the screen, and the average value of the instantaneous loudness value after the current time is the remaining time target. By adjusting the audio level so that it is below the allowable upper limit value of the average loudness value L ₀ and above the allowable lower limit value, the average loudness value of the entire program at the end of the measurement can be easily set as the target loudness value L _t. You can get closer.

及び有効ブロック数ｍを、見込み有効ブロック数算出手段１２から見込み有効ブロック数Ｎをそれぞれ入力する。そして、超過・不足ラウドネス値算出手段１５は、これらの値に基づいて、補助指標である超過ラウドネス値ＥＬＬ及び不足ラウドネス値ＳＬＬを算出する。超過・不足ラウドネス値算出手段１５により算出された超過ラウドネス値ＥＬＬ及び不足ラウドネス値ＳＬＬは、超過・不足ラウドネス値表示手段１６に出力される。 [Excess / insufficient loudness value calculation means 15]
Returning to FIG. 3, the excess / insufficient loudness value calculation means 15 inputs the preset target loudness value L _t , the allowable upper limit loudness value _Lu and the allowable lower limit loudness value L _l , and the average loudness value calculation means 10. To root mean square

And the number m of effective blocks are input as the number N of expected effective blocks from the expected effective block number calculation means 12. Then, the excess / insufficient loudness value calculation means 15 calculates an excess loudness value ELL and an insufficient loudness value SLL, which are auxiliary indicators, based on these values. The excess loudness value ELL and the insufficient loudness value SLL calculated by the excess / insufficient loudness value calculation means 15 are output to the excess / insufficient loudness value display means 16.

  FIG. 9 is a block diagram showing the configuration of the excess / insufficient loudness value calculating means 15, and FIG. 10 is a flowchart showing the processing of the excess / insufficient loudness value calculating means 15. The excess / insufficient loudness value calculation means 15 includes a loudness value / root mean square conversion stage 30, a memory 31, and an excess / insufficient loudness value calculation stage 32.

The loudness value / root mean square conversion stage 30 inputs a preset target loudness value L _t , an allowable upper limit loudness value _Lu and an allowable lower limit loudness value L _l (step S1001), and uses the above equation (10). These values are converted into the target mean square value Z _t , the allowable upper limit mean square value Z _u, and the permissible lower limit mean square value Z _l (step S1002) and stored in the memory 31.

及び有効ブロック数ｍを、見込み有効ブロック数算出手段１２から見込み有効ブロック数Ｎを入力し（ステップＳ１００３）、これらの値をメモリ３１に保存する。 The excess / insufficient loudness value calculation means 15 is the mean square value from the average loudness value calculation means 10.

Then, the effective block number m is input as the expected effective block number N from the expected effective block number calculating means 12 (step S1003), and these values are stored in the memory 31.

、有効ブロック数ｍ及び見込み有効ブロック数Ｎを読み出し、二乗平均値

とターゲット二乗平均値Ｚ_tとを比較する（ステップＳ１００４）。 The excess / insufficient loudness value calculation stage 32 reads the target mean square value Z _t , the allowable upper limit mean square value Z _u , the permissible lower limit mean square value Z _l , and the mean square value from the memory 31.

Read out the effective block number m and the expected effective block number N, and the mean square value

And comparing the target mean square value Z _t (step S1004).

がターゲット二乗平均値Ｚ_t以上であると判定した場合、前記式（８）により、超過ラウドネス値ＥＬＬを算出し（ステップＳ１００５）、前記式（９）により、不足ラウドネス値ＳＬＬに−∞を設定する（ステップＳ１００６）。 In step S1004, the excess / insufficient loudness value calculation stage 32 calculates the root mean square value.

Setting but if it is determined that the target mean square value Z _t or more, according to the equation (8), by calculating the excess loudness value ELL (step S1005), the formula (9), the -∞ to lack loudness value SLL (Step S1006).

がターゲット二乗平均値Ｚ_tよりも小さいと判定した場合、前記式（８）により、超過ラウドネス値ＥＬＬに−∞を設定し（ステップＳ１００７）、前記式（９）により、不足ラウドネス値ＳＬＬを算出する（ステップＳ１００８）。 On the other hand, the excess / insufficient loudness value calculation stage 32 determines the root mean square value in step S1004.

Calculating case but it is determined to be smaller than the target mean square value Z _t, by the formula (8), by setting the -∞ to excess loudness value ELL (step S1007), the formula (9), the lack of loudness values SLL (Step S1008).

  The excess loudness value ELL and the insufficient loudness value SLL calculated or set by the excess / insufficient loudness value calculation stage 32 are output to the excess / insufficient loudness value display means 16 (step S1009).

[Excess / insufficient loudness value display means 16]
Returning to FIG. 3, the excess / insufficient loudness value display means 16 inputs the excess loudness value ELL and the insufficient loudness value SLL from the excess / insufficient loudness value calculation means 15 and displays these values on the screen.

FIG. 11 is a diagram showing an example of a display form of the excess loudness value ELL and the insufficient loudness value SLL displayed on the screen by the excess / insufficient loudness value display means 16. The excess loudness value ELL is defined to have a finite value only if the current average loudness value L _k is greater than the target loudness value L _t , and the insufficient loudness value SLL is the current average loudness value S L It is defined to have a finite value only if L _k is smaller than the target loudness value L _t . When one of the excess loudness value ELL or the insufficient loudness value SLL has a value, the other is −∞ (dB).

The display form of the excess loudness value ELL and the insufficient loudness value SLL shown in FIG. 11 is such that these levels are displayed in one level meter. The level of the excess loudness value ELL is displayed on the upper part between the center and the upper end of the level meter (the side indicating that it is exceeded), and the level of the insufficient loudness value SLL is between the center and the lower end of the level meter. Is displayed at the bottom (the side indicating that it is missing). The center of the level meter is −∞ (dB), and when the average loudness value L _k at the current time coincides with the target loudness value L _t , the level is displayed at the position of the center of the level meter. When the average loudness value L _k is larger than the target loudness value L _t , the level swings upward from the center position. Conversely, when the level is insufficient, the level swings downward from the center position.

When the average loudness value L _k is larger than the allowable upper limit loudness value L _u , the level meter swings upward from the upper 0 (dB) position. By displaying this excess portion in red or the like (displayed in a display form different from the case where the average loudness value L _k is larger than the target loudness value L _{t and} less than the allowable upper limit loudness value L _u ), the program producer Can call attention. When the average loudness value L _k is smaller than the allowable lower limit loudness value L _l , the level meter swings downward from the lower 0 (dB) position. By displaying this lacking portion in red or the like (displayed in a display form different from the case where the average loudness value L _k is smaller than the target loudness value L _t and is equal to or greater than the allowable lower limit loudness value L _l ), the program producer Can call attention.

がターゲット二乗平均値Ｚ_t以上であり、超過ラウドネス値ＥＬＬが−∞より大きく、かつ０より小さい場合の表示形態を示している。番組制作者は、図１１（ａ）に示す画面表示により、超過ラウドネス値ＥＬＬの程度を知ることができ、平均ラウドネス値Ｌ_kが許容上限ラウドネス値Ｌ_uに達するまで余裕があることを認識することができる。 FIG. 11A shows the root mean square value.

Shows a display form in which is _{equal to} or greater than the target root mean square value Z _t and the excess loudness value ELL is greater than −∞ and less than 0. Program producer will recognize that the screen display shown in FIG. 11 (a), it is possible to know the degree of excess loudness values ELL, there is a margin to a mean loudness value L _k reaches the allowable upper limit loudness value L _u be able to.

がターゲット二乗平均値Ｚ_t以上であり、超過ラウドネス値ＥＬＬが０以上である場合の表示形態を示している。番組制作者は、図１１（ｂ）に示す画面表示により、超過ラウドネス値ＥＬＬの程度を知ることができ、平均ラウドネス値Ｌ_kが許容上限ラウドネス値Ｌ_uに達しており、超過していることを認識することができる。 FIG. 11B shows the mean square value

There is a target mean square value Z _t or more, show the display form when excess loudness value ELL is greater than zero. Program producer is the screen display shown in FIG. 11 (b), it is possible to know the degree of excess loudness value ELL, average loudness value L _k has reached the allowable upper limit loudness value L _u, it has been exceeded Can be recognized.

がターゲット二乗平均値Ｚ_tより小さく、不足ラウドネス値ＳＬＬが−∞より大きく、かつ０より小さい場合の表示形態を示している。番組制作者は、図１１（ｃ）に示す画面表示により、不足ラウドネス値ＳＬＬの程度を知ることができ、平均ラウドネス値Ｌ_kが許容下限ラウドネス値Ｌ_lに達するまで余裕があることを認識することができる。 FIG. 11C shows the root mean square value.

Shows a display form in which is smaller than the target mean square value Z _t , the insufficient loudness value SLL is larger than −∞, and smaller than 0. The program producer can know the degree of the insufficient loudness value SLL from the screen display shown in FIG. 11C, and recognizes that there is a margin until the average loudness value L _k reaches the allowable lower limit loudness value L _l. be able to.

がターゲット二乗平均値Ｚ_tより小さく、不足ラウドネス値ＳＬＬが０以上である場合の表示形態を示している。番組制作者は、図１１（ｄ）に示す画面表示により、不足ラウドネス値ＳＬＬの程度を知ることができ、平均ラウドネス値Ｌ_kが許容下限ラウドネス値Ｌ_lに達しており、不足していることを認識することができる。 FIG. 11 (d) shows the root mean square value

There smaller than the target mean square value Z _t, show the display form when insufficient loudness value SLL is greater than zero. The program producer can know the degree of the insufficient loudness value SLL from the screen display shown in FIG. 11D, and the average loudness value L _k has reached the allowable lower limit loudness value L _l and is insufficient. Can be recognized.

11A to 11D, the program producer can intuitively know the excess loudness value ELL and the insufficient loudness value SLL, and the average loudness value _Lk is exceeded. Recognize the degree of deficiency and margin. Accordingly, program producer is the average loudness value of the whole program in the measuring end can more easily close to meeting the target loudness value L _t.

、現在時点の平均ラウドネス値Ｌ_k、及び破棄されなかった二乗平均値Ｚ_iの数である有効ブロック数ｍを算出するようにした。 As described above, according to the sound level monitoring apparatus 1 of the embodiment of the present invention, the average loudness value calculating unit 10 calculates the mean square value Z _k of the sound signal for each block by the mean square value calculating process for each block. calculated by the absolute gating process, using the mean square value Z _k, it discards the mean square value Z _k that can be regarded as a predetermined size following the silence by threshold processing, the relative gating process, the absolute gating process using mean square value Z _i which have not been discarded Te, discards the mean square value Z _i that can be regarded as a predetermined size or less of the small sound by thresholding, using a not discarded mean square value Z _i, the current Root mean square value

The effective block number m, which is the number of the average loudness value L _k at the current time point and the mean square value Z _i that has not been discarded, is calculated.

、有効ブロック数ｍ及び見込み有効ブロック数Ｎを用いて、前記（２）及び前記（３）により、補助指標である残時間目標平均ラウドネス値Ｌ₀を算出する。同様に、残時間目標平均ラウドネス値算出手段１３は、許容上限ラウドネス値Ｌ_uに対する残時間目標平均ラウドネス値Ｌ₀の許容上限値、及び、許容下限ラウドネス値Ｌ_lに対する残時間目標平均ラウドネス値Ｌ₀の許容下限値を補助指標として算出する。 Then, the expected effective block number calculating means 12 calculates the expected effective block number N of the entire program at the current time point by the above equation (1) using the preset measurement time length N _all and the effective block number m. I did it. In addition, the remaining time target average loudness value calculation means 13 calculates the target mean square value Z _t corresponding to the preset target loudness value L _t , the mean square value at the current time point.

The remaining time target average loudness value L ₀ , which is an auxiliary index, is calculated using (2) and (3) using the effective block number m and the expected effective block number N. Similarly, the remaining time target average loudness value calculating means 13 is a permissible upper limit value of the remaining time target average loudness value L _{0 for} the allowable upper limit loudness value L _{u and} a remaining time target average loudness value L for the allowable lower limit loudness value L _l . An allowable lower limit value of ₀ is calculated as an auxiliary index.

The remaining time target average loudness value L ₀ , which is an auxiliary index, is an index that takes into account the remaining time. By matching the average loudness value from the current time point to the measurement end time to this value, the entire program at the time of the measurement end time Is the value that makes it possible to bring the average loudness value close to the target loudness value L _t . However, in practice, the average loudness value from the current time point to the measurement end time point is a future value after the current time point, and therefore can only be predicted from instantaneous fluctuations in the loudness value. Accordingly, program producer, the residue refers to the time target average loudness value L _0, the average value of the instantaneous loudness value to adjust the audio level to match the remaining time target average loudness value L ₀ Thus, the average loudness value of the entire program can be easily brought close to the target loudness value L _t without regard to the elapsed time from the measurement start time.

Allowable upper limit and the allowable lower limit value of the remaining time as an auxiliary indicator target average loudness value L ₀ is an index that takes into account the remaining time, an average loudness value from the current time to the measurement end, below the allowable upper limit The average loudness value of the entire program at the end of measurement is brought close to the allowable upper limit loudness value L _u and lower than the allowable lower limit loudness value L _{l by} setting the value to the allowable lower limit value or more. This is a possible value. However, in practice, the average loudness value from the current time point to the measurement end time point is a future value after the current time point, and therefore can only be predicted from instantaneous fluctuations in the loudness value. Therefore, the program producer refers to the allowable upper limit value and the allowable lower limit value of the remaining time target average loudness value L ₀ , and the average value of the instantaneous loudness value is the allowable value of the remaining time target average loudness value L ₀ . By adjusting the audio level so that it is less than the upper limit and greater than the allowable lower limit, the average loudness value of the entire program can be simply set to the target loudness value L _t without regard to the elapsed time from the start of measurement. You can get closer.

、有効ブロック数ｍ及び見込み有効ブロック数Ｎを用いて、前記（８）及び前記（９）により、補助指標である超過ラウドネス値ＥＬＬ及び不足ラウドネス値ＳＬＬを算出するようにした。 Furthermore, the excess-deficient loudness value calculation unit 15, a target mean square value Z _t corresponding to a preset target loudness value L _t, allowable upper mean square value Z corresponding to the allowable upper loudness value L _u that is set in advance _u , allowable lower limit loudness value L _l corresponding to preset allowable lower limit loudness value L _l , mean square value at present

The excess loudness value ELL and the insufficient loudness value SLL, which are auxiliary indexes, are calculated by the above (8) and (9) using the effective block number m and the expected effective block number N.

The excess loudness value ELL, which is an auxiliary indicator, is an indicator that takes into account the remaining time, and is a value that indicates the degree of excess and margin of the average loudness value L _k at the current time point. The insufficient loudness value SLL, which is an auxiliary indicator, It is an index that takes into account the remaining time, and is a value that represents the shortage and margin of the average loudness value L _k at the current time point. Therefore, the program producer refers to the excess loudness value ELL and the insufficient loudness value SLL, and determines the average loudness value L _k of the entire program to be measured without regard to the elapsed time from the measurement start time. it can be conveniently close to the value L _t.

  Note that a normal computer can be used as the hardware configuration of the sound level monitoring apparatus 1 according to the embodiment of the present invention. The sound level monitoring apparatus 1 is configured by a computer including a volatile storage medium such as a CPU and a RAM, a non-volatile storage medium such as a ROM, an interface, and the like. Average loudness value calculation means 10, average loudness value display means 11, expected effective block number calculation means 12, remaining time target average loudness value calculation means 13, remaining time target average loudness value display means 14 provided in the sound level monitoring device 1, The functions of the excess / insufficient loudness value calculation means 15 and the excess / insufficient loudness value display means 16 are each realized by causing the CPU to execute a program describing these functions. These programs are stored in the storage medium and read out and executed by the CPU. These programs can also be stored and distributed in a storage medium such as a magnetic disk (floppy (registered trademark) disk, hard disk, etc.), optical disk (CD-ROM, DVD, etc.), semiconductor memory, etc. You can also send and receive.

The present invention has been described with reference to the embodiment. However, the present invention is not limited to the above-described embodiment, and various modifications can be made without departing from the technical idea thereof. For example, in the above embodiment, the remaining time target average loudness value calculation means 13 uses the expected effective block number N and the measured time length N _all that is the number of blocks of the entire program, etc., and the remaining time target average loudness value L _0. Was calculated. In this example, the program time length is fixed, and the measurement time length N _all which is the total number of blocks is set to a fixed time length.

On the other hand, the remaining time target average loudness value L is calculated by the remaining time target average loudness value calculation means 13 even when the program time length is not fixed but variable, and the measurement time length N _all which is the total number of blocks is not set. ₀ is calculated. Specifically, assuming a measurement time length N _all that is the number of blocks of the entire program, a fixed block number having a sufficiently short time length with respect to the measurement time length N _all is set in advance. The remaining time target average loudness value calculating means 13 matches the average loudness value L _k at a time (shifted with the passage of time) after a fixed number of blocks from the current time point with the target loudness value L _t with reference to the current time point. As described above, the remaining time target average loudness value L ₀ may be calculated by the above equations (2) and (3). In this case, in the above formula (1), the expected effective block number N is calculated assuming that the measurement time length N _all is the number of blocks from the measurement start time to the time after the fixed block number from the current time. . In addition, when the end of the program is set during the calculation process, the fixed block number is determined from the timing when the remaining block number from the current time point to the measurement end point which is the end of the program is smaller than the fixed block number. Or you may make it count down with progress of time from the preset timing.

As a result, even if the program time length is not fixed but variable, and the end of the program is not set in advance, the remaining time target average loudness value L ₀ that is an auxiliary index can be calculated.

Further, the remaining time target average loudness value calculating means 13 sets the measurement end time at the time after the measurement time length N _all which is the number of blocks of the entire program from the measurement start time to the measurement end time at the measurement start time. Set. Thereby, measurement time length can be made into fixed time length. When the remaining time target average loudness value calculation means 13 inputs a switching signal in accordance with the operation of the operator, the measurement end time is changed by setting the measurement end time to a time after a predetermined time from the current time. Thereby, the measurement time length can be made a variable time length. Further, when the remaining time target average loudness value calculation means 13 inputs an end signal in accordance with the operation of the operator, the remaining time target average loudness value calculation means 13 stops the process of setting the measurement end time to a time after a predetermined time from the current time, and the already set measurement end time Is fixed.

If the program configuration is determined in advance, target loudness values L _{t ′} on the target program configuration are set at a plurality of points in the middle of the program, and each time point is set as a measurement end point. It may be considered. The target loudness value L _{t ′} on the program structure at a plurality of points in the middle of the program is set so that the average loudness value L _{k of the} entire program matches the target loudness value L _t . When the program is configured to lower the audio level in the first half of the program and increase the audio level in the second half of the program, it is assumed that the average loudness value L _{k of the} entire program matches the target loudness value L _t = −24LKFS. , for example, the target loudness value on the program structure of the program half L _t '= -24.5LKFS, and target loudness value L _t on the program configuration of late program' = -23.5LKFS is preset.

In this case, the expected effective block number calculating means 12 sets the measurement time length N _all in the equation (1) as the time length of the number of blocks from the measurement start time to each time set during the program, The number N of expected effective blocks is calculated. Then, the remaining time target average loudness value calculating means 13 uses the target loudness value L _{t ′} on the program structure at the most recent time among a plurality of time points set during the program at the current time point, to determine the remaining time target average value. A loudness value L ₀ is calculated.

As a result, the program producer sets the most recent time among a plurality of time points set in the middle of the program as the measurement end time point, and the remaining time target average loudness value L ₀ corresponding to the target loudness value L _{t ′} on the program structure. Can be referred to as an auxiliary indicator, and the average loudness value L _k of the entire program can be easily brought close to the target loudness value L _t according to the configuration of the program.

In the above embodiment, the average loudness value calculating means 10, relative the gating process, but to perform the thresholding loudness value L _a calculated in absolute gating process with reference loudness value L _a it may perform threshold processing on the basis of the target loudness value L _t in place of. Accordingly, since it is unnecessary to calculate the loudness value L _a at absolute gating process without calculating the relative gating treated with absolute gating processes similar calculation flow, it is possible to reduce the processing load .

  In the above embodiment, the excess / deficiency loudness value display means 16 displays the excess loudness value ELL and the insufficient loudness value SLL on the screen in the display forms shown in FIGS. 11 (a) to 11 (d). However, this display form is an example. The present invention is not limited to this display form, and any display form that allows a user such as a program producer to intuitively recognize the degree of loudness value excess, deficiency, and margin.

  The sound level monitoring apparatus for calculating an auxiliary index according to the present invention is useful in the field of managing sound levels based on the loudness value of a television program.

DESCRIPTION OF SYMBOLS 1 Voice level monitoring apparatus 10 Average loudness value calculation means 11 Average loudness value display means 12 Expected effective block number calculation means 13 Remaining time target average loudness value calculation means 14 Remaining time target average loudness value display means 15 Excess / insufficient loudness value calculation means 16 Excess / insufficient loudness value display means 20, 30 Loudness value / root mean square conversion stage 21, 31 Memory 22 Remaining time target mean square value calculation stage 23 Mean square value / loudness value conversion stage 32 Excess / insufficient loudness value calculation stage

Claims (8)

In an audio level monitoring device for monitoring the audio level of a program,
An average loudness value calculating means for inputting an audio signal of the program and calculating an average value of a loudness value between a predetermined measurement start time and a current time as an average loudness value at the current time;
The average loudness value at the current time point calculated by the average loudness value calculating means, the target loudness value indicating the target value of the average loudness value of the entire program, and the measurement from the measurement start time to a predetermined measurement end time Data for approximating the average loudness value at the measurement end time calculated by the average loudness value calculation means in the remaining time from the current time point to the measurement end time based on the time length to the target loudness value An auxiliary index calculating means for calculating as an auxiliary index for adjusting the sound level;
An audio level monitoring device comprising: The voice level monitoring apparatus according to claim 1,
The auxiliary index calculating means includes
The average loudness value at the current time point calculated by the average loudness value calculating means, the target loudness value indicating the target value of the average loudness value of the entire program, and the measurement from the measurement start time to a predetermined measurement end time Based on the time length, the remaining time target average loudness value that can be the target of the average loudness value in the remaining time from the current time point to the measurement end time point is calculated as the auxiliary indicator. Monitoring device. The voice level monitoring apparatus according to claim 1,
The auxiliary index calculating means includes
The average loudness value calculated by the average loudness value calculating means, the allowable upper limit loudness value indicating the allowable upper limit of the average loudness value of the entire program, and the period from the measurement start time to a predetermined measurement end time Calculating an allowable upper limit value of the remaining time target average loudness value that can be an allowable upper limit of the average loudness value in the remaining time from the current time point to the measurement end time point, based on the measurement time length, as the auxiliary index; A voice level monitoring device characterized by the above. The voice level monitoring apparatus according to claim 1,
The auxiliary index calculating means includes
The average loudness value at the present time calculated by the average loudness value calculating means, an allowable lower limit loudness value indicating an allowable lower limit of the average loudness value of the entire program, and a period from the measurement start time to a predetermined measurement end time Calculating an allowable lower limit value of a remaining time target average loudness value that can be an allowable lower limit of an average loudness value in a remaining time from the current time point to the measurement end time point, as the auxiliary index, based on a measurement time length. A voice level monitoring device characterized by the above. The voice level monitoring apparatus according to claim 1,
The auxiliary index calculating means includes
An excess allowable limit indicating a difference between the target loudness value and an allowable upper limit loudness value indicating an allowable upper limit of the average loudness value of the entire program in a measurement time length from the measurement start time to the measurement end time; The difference between the average loudness value and the target loudness value when the average loudness value calculated by the average loudness value calculation means is greater than or equal to the target loudness value in the time length from the measurement start time to the current time. Calculating the ratio between the excess amount indicating the excess loudness value,
Insufficient allowable amount indicating the difference between the target loudness value and the allowable lower limit loudness value indicating the allowable lower limit of the average loudness value of the entire program in the measurement time length, and between the measurement start time and the current time In the time length, the ratio between the average loudness value calculated by the average loudness value calculation means is equal to or less than the target loudness value and the shortage indicating the difference between the target loudness values is insufficient. Calculated as the loudness value,
The voice level monitoring apparatus, wherein the excess loudness value and the insufficient loudness value are used as the auxiliary index. The voice level monitoring apparatus according to claim 5, wherein
Excessive / insufficient loudness values displayed by a single level meter having an excess loudness value and an insufficient loudness value calculated by the auxiliary index calculating means and a side indicating an excess. A display means,
The excess / insufficient loudness value display means is:
If the current average loudness value matches the target loudness value, display the level in the center of the level meter;
If the average loudness value at the current time point is greater than the target loudness value and less than or equal to the allowable upper limit loudness value, the level is displayed in a first form on the side indicating the excess in the level meter, If the average loudness value at the current time point is greater than the target loudness value and greater than the allowable upper limit loudness value, the level is different from the first level on the side indicating the excess in the level meter. Is displayed,
If the current average loudness value is smaller than the target loudness value and greater than or equal to the allowable lower limit loudness value, a level is displayed in a second form on the side indicating the lack in the level meter, When the average loudness value at the current time point is smaller than the target loudness value and smaller than the allowable lower limit loudness value, the second meter is different from the second on the side indicating the lack in the level meter. An audio level monitoring apparatus characterized by displaying a level. In the voice level monitoring device according to any one of claims 1 to 6,
At the measurement start time, the measurement time length from the measurement start time to the measurement end time is set to a fixed time length, and when a switching signal is input during measurement, the measurement end time is set to be more than the current time point. An audio level monitoring apparatus, characterized in that a measurement time length from a measurement start time to a measurement end time is set to a variable time length after a predetermined time.   The program for functioning a computer as an audio | voice level monitoring apparatus as described in any one of Claim 1-7.

Images