Loudness measuring device and program

Abstract

To determine a loudness value further in line with human hearing.SOLUTION: A characteristic filter processing unit includes a first filter processing unit that inputs multi-channel acoustic signals and performs first filter processing to add predetermined spatial acoustic characteristics to the acoustic signals of each channel, and a second filter processing unit that performs second filter processing to add predetermined auditory characteristics to the acoustic signals input from the first filter processing unit. A loudness calculation unit determines a loudness value based on intensity obtained by averaging power of the acoustic signals input from the characteristic filter processing unit, by using weight, across channels, and a filter control unit determines whether or not to perform the first filter processing depending on whether the multi-channel acoustic signals are binaural signals.SELECTED DRAWING: Figure 1

Description

[Technical Field]

[0001] The present invention relates to loudness measurement devices and programs, for example, techniques for determining loudness in binaural reproduction of audio signals. [Background technology]

[0002] Binaural reproduction technology reproduces the sound pressure generated when sound waves emitted from a sound source placed at any position reach the entrances of the ear canals of both ears. Binaural reproduction uses playback sources, such as headphones or earphones, placed close to the left and right ears to present sound. When reproducing the sound pressure from a sound source, the transmission characteristics of the sound from the sound source to each ear include the effects of reflection, diffraction, and attenuation of sound waves from the listener's head, pinna, torso, etc. (hereinafter referred to as "torso, etc."). Binaural reproduction technology can be divided into two methods: one that uses microphones placed at the entrances of the ear canals of a dummy head that mimics the shape of a human head, and one that reproduces sound by adding pre-measured transmission characteristics from the sound source to each ear. The head-related transfer function (HRTF) or head-related impulse response (HRIR) is used to represent the transmission characteristics. HRTF is a function that represents the transfer characteristics of sound from the sound source to the entrances of the ear canals of both ears in the frequency domain. When adding transfer characteristics using HRTF, the HRTF is multiplied by the conversion coefficient for each frequency obtained by converting the acoustic signal to the frequency domain. HRIR is a response coefficient that represents the transfer characteristics of sound from the sound source to the entrances of the ear canals of both ears in the time domain. HRIR is equivalent to an impulse response that expresses HRTF in the time domain. When adding transfer characteristics using HRIR, a convolution operation is performed with the HRIR on the acoustic signal, which is a time series of sample values ​​in the time domain.

[0003] In recent years, multi-channel audio has become widespread, while the usage pattern of individuals using mobile information devices to view desired content has become established. For example, 5.1 surround (also called "5.1ch audio") and 22.2 multi-channel audio (also called "22.2ch audio") have been proposed as multi-channel audio (Non-Patent Document 1). Information processing devices that are not necessarily primarily intended for audio playback, such as smartphones and tablet devices, are often used as mobile information devices. Under these circumstances, binaural playback technology is useful as a means for enjoying simulated multi-channel audio content.

[0004] In the broadcasting field, standardization of audio processing methods that include binaural playback technology is underway. For example, binaural playback technology is defined as one of the playback formats in the Audio Definition Model (ADM) described in Non-Patent Document 2 and the audio coding method MPEG-H 3DA described in Non-Patent Document 3. ADM is audio metadata, or its format, for describing playback format information as well as program audio material information.

[0005] In addition to differences in sound reproduction technology, variations in volume can occur among individually produced audio content. To unify the volume across multiple pieces of content, program-wide volume management based on loudness values ​​is sometimes implemented. The loudness value is an index value that represents psychological loudness. For example, Patent Documents 1-3 propose methods for measuring the loudness value of content in various playback formats, such as monaural playback, stereo playback, and multi-channel audio. Non-Patent Document 4 also standardizes loudness value measurement methods for multi-channel audio (hereinafter, existing loudness value measurement methods are collectively referred to as "conventional methods") and target values. As illustrated in Figure 4, in the conventional method, the input audio signal of each channel is filtered using a K-weighting filter. The K-weighting filter is a two-stage pre-filter, and the first-stage pre-filter (first pre-filter) primarily simulates the physical effect of the head on sound propagation. The second-stage prefilter (second prefilter) is a filter that adds Revised Low-frequency B-weighting (RLB) to simulate human hearing characteristics. The weighted average value is obtained by weighting the root mean square of the audio signal obtained by the K-weighting filter processing using a weighting coefficient set in advance for each channel. The obtained weighted average value is converted to the logarithmic domain, and the loudness value is calculated by truncating the converted values ​​in the gating process that are below a certain loudness threshold. [Prior art documents] [Patent documents]

[0006] [Patent Document 1] Patent No. 3462390 [Patent Document 2] Japanese Patent Application Laid-Open No. 2006-279242 [Patent Document 3] Patent No. 6510757 [Non-patent literature]

[0007] [Non-Patent Document 1] Rec. ITU-R BS.2051-2 “Advanced sound system for program production” (2018) [Non-patent document 2] Rec. ITU-R BS.2076-2 “Audio definition model” (2019) [Non-patent document 3] ISO / IEC 23008-3:2019 “Information technology - High efficiency coding and media delivery in heterogeneous environments - Part 3: 3D audio, Second edition” (2019) [Non-patent document 4] Rec. ITU-R BS.1770-4 “Algorithms to measure audio program loudness and true-peak audio level” (2015) Summary of the Invention [Problem to be solved by the invention]

[0008] Although loudness measurement methods for multichannel audio content have been standardized, methods for measuring loudness for binaurally played content have not been clearly defined. Binaural signals (such as those recorded using the artificial head described above and those with sound transfer characteristics added using HRIRs or HRTFs) are a type of multi-channel audio signal, and so conventional methods could be applied. However, binaural signals reflect physical acoustic effects such as the reflection and diffraction of sound waves in the human head. If conventional methods are used as is, the acoustic effects of K-weighting filtering will be redundantly included. Furthermore, the frequency characteristics specific to the equipment used to measure HRIRs or HRTFs may overlap with the corrections for auditory characteristics. Therefore, conventional methods do not necessarily produce loudness values ​​that are consistent with human hearing.

[0009] The present invention has been made to solve the above-mentioned problems, and one object of the present invention is to provide a loudness value measuring device and program that can obtain loudness values ​​that are more in line with human hearing. [Means for solving the problem]

[0010] [1] In one aspect of the present invention, a multi-channel acoustic signal is input, and the acoustic signal of each channel is subjected to a This is a filtering process that uses a filter that simulates the acoustic effects of the human head. a characteristic filter processing unit including a first filter processing unit that performs first filter processing and a second filter processing unit that performs second filter processing to add predetermined auditory characteristics to the audio signal input from the first filter processing unit; a loudness calculation unit that determines a loudness value based on an intensity obtained by weighted averaging the power of the audio signal input from the characteristic filter processing unit across channels; and the audio signals of the multiple channels are binaural signals. case , performing the first filtering process and if the multi-channel acoustic signal is not a binaural signal, perform the first filtering process. and a filter control unit that determines the loudness. According to the configuration of [1], when a binaural signal is input to the characteristic filter processing unit, the first filter processing is not performed, so the spatial acoustic characteristics included in the binaural signal and the spatial acoustic characteristics resulting from the first filter processing do not overlap. As a result, a loudness value that conforms to the auditory sensation can be obtained. Furthermore, when a binaural signal is not input to the characteristic filter processing unit, the first filter processing is performed, so that the requirements for existing loudness measurements can be met.

[0011] [2] One aspect of the present invention is Equipped with a binaural signal generator In the loudness measuring device described above, the binaural signals include a left-ear signal for a sound to be presented to the left ear and a right-ear signal for a sound to be presented to the right ear, The binaural signal generation unit A signal for the left ear is generated by adding a transfer characteristic from the sound source to the left ear to the sound source signal, and a signal for the right ear is generated by adding a transfer characteristic from the sound source to the sound source signal to the right ear. death That's fine. According to the configuration of [2], a binaural signal is generated based on a sound source signal, and a loudness value for the generated binaural signal is obtained. Therefore, the obtained loudness value can be used to efficiently edit or produce content that includes the binaural signal.

[0012] [3] In one aspect of the present invention, in the loudness measuring device described above, the filter control unit at a low frequency equal to or lower than a predetermined reference frequency Transmission Whether or not to perform the second filtering process may be determined depending on whether or not the gain of the characteristic is equal to or greater than a predetermined reference gain. According to the configuration of [3], the second filter processing is performed when the gain of the transfer characteristic in the low frequency range is high, so that a loudness value that matches the auditory sense can be obtained, taking into account the small contribution to the auditory sense in the low frequency range.

[0013] [4] One aspect of the present invention is a computer of, It may also be a program for causing the loudness measurement device to function as the above-mentioned loudness measurement device. According to the configuration of [4], when a binaural signal is input to the characteristic filter processing unit, the first filter processing is not performed, and therefore the spatial acoustic characteristics included in the binaural signal and the spatial acoustic characteristics resulting from the first filter processing do not overlap, resulting in a loudness value that matches the auditory sensation. Furthermore, when a binaural signal is not input to the characteristic filter processing unit, the first filter processing is performed, and therefore the requirements for existing loudness measurements can be met. [Effects of the Invention]

[0014] According to the present invention, a loudness value that is more in line with human hearing can be obtained. [Brief explanation of the drawings]

[0015] [Figure 1] 1 is a schematic block diagram showing an example of the functional configuration of a loudness measurement device according to an embodiment of the present invention. [Figure 2]FIG. 10 is a schematic block diagram showing another example of the functional configuration of the loudness measurement device according to the present embodiment. [Figure 3] FIG. 2 is a schematic block diagram illustrating an example of the functional configuration of a convolution calculation unit according to the present embodiment. [Figure 4] FIG. 1 is a schematic block diagram showing an example of an existing loudness measurement method. DETAILED DESCRIPTION OF THE INVENTION

[0016] Hereinafter, an embodiment of the present invention will be described with reference to the drawings. First, an example of the functional configuration of a loudness measurement device 10 according to this embodiment will be described. Fig. 1 is a schematic block diagram showing an example of the functional configuration of a loudness measurement device 10 according to this embodiment. The loudness measuring device 10 includes a characteristic filter processing unit 110, a loudness calculation unit 120, and a filter control unit 140. In the example of FIG. 1, there are two characteristic filter processing units 110. This number corresponds to the number of channels of the input acoustic signal. The individual characteristic filter processing units 110 and their constituent parts are distinguished by sub-numbers such as 110-1, 110-2, etc. However, in the present application, sub-numbers may not be specifically assigned to matters common to multiple characteristic filter processing units 110.

[0017] Each characteristic filter processing unit 110 performs filtering on the acoustic signal x of the corresponding channel. The characteristic filter processing unit 110 is capable of performing first filtering and second filtering on the acoustic signal of each channel. In addition, the characteristic filter processing unit 110 can change whether or not to perform the first filtering and the second filtering. In the example of FIG. 1, the characteristic filter processing unit 110 includes two switch units 112 and 116, a first filtering unit 114, and a second filtering unit 118.

[0018] The switch unit 112 controls whether to output the acoustic signal input to the switch unit 112 to the first filter processing unit 114 or to the switch unit 116, bypassing (successfully passing) the first filter processing unit 114, based on a first control signal input from the filter control unit 140. Whether to bypass the first filter processing unit 114 is determined by whether or not the first filter processing is required. When the first control signal indicating that the first filter processing is required is input, the switch unit 112 outputs the acoustic signal input to the switch unit 112 to the first filter processing unit 114. When the second control signal indicating that the first filter processing is not required is input, the switch unit 112 outputs the acoustic signal input to the switch unit 112 to the second filter processing unit 118.

[0019] The first filtering unit 114 performs first pre-filtering on the acoustic signal input from the switch unit 112 and outputs the resulting acoustic signal as a first filtered signal to the switch unit 116. The first pre-filtering may be similar to the first stage of pre-filtering of a K-weighting filter (K-filter) described in Non-Patent Document 4. The first pre-filter simulates the acoustic effect produced by modeling a human head as a rigid sphere. The first pre-filter functions as a shelving filter, in which the gain increases as the frequency increases, but the gain approaches a first reference value as the frequency approaches zero and a second reference value as the frequency approaches infinity. The second reference value is a positive real number greater than the first reference value. The first pre-filter is implemented, for example, as an auto-regressive moving average (ARMA) filter.

[0020] Based on a second control signal input from the filter control unit 140, the switch unit 116 controls whether the output destination of the acoustic signal input from the switch unit 112 or the first filtered signal input from the first filtering unit 114 is to the second filtering unit 118 or to the loudness calculation unit 120 by bypassing the second filtering unit 118. Whether or not to perform second filtering is determined by whether or not to bypass the second filtering unit 118. When the second control signal indicating that second filtering is required is input, the switch unit 116 outputs the acoustic signal input to the switch unit 116 to the second filtering unit 118. When the second control signal indicating that second filtering is not required is input, the switch unit 116 outputs the acoustic signal input to the switch unit 116 as an output signal y to the loudness calculation unit 120.

[0021] The second filtering unit 118 performs second pre-filtering on the audio signal input from the switch unit 116, and outputs the resulting audio signal as an output signal y to the loudness calculation unit 120. The second filtering may be performed using different parameters under a configuration similar to that of the second stage of pre-filtering of the K-weighting filter described in Non-Patent Document 4. The second pre-filter may also be implemented as, for example, an ARMA filter. The second pre-filter functions as an auditory weighting filter that simulates the human auditory characteristics. The second pre-filter functions as a high-pass filter whose gain increases as the frequency increases, but gradually approaches a certain reference value.

[0022] The loudness calculation unit 120 calculates a loudness value from the output signal y input from the characteristic filter processing unit 110 corresponding to each channel. The loudness calculation unit 120 includes a root mean square unit 122, a weighting unit 124, an adder 126, a decibel conversion unit 128, and a gate processing unit 130. In the loudness calculation unit 120, the root mean square unit 122 and the weighting unit 124 are provided for each channel, and process the signal of each channel or a numerical value obtained from that signal.

[0023] The mean square unit 122 calculates the mean square z of the signal values ​​of each sample that constitutes the output signal y for each gating block (extraction section) that is shorter than a predetermined period T. Adjacent gating blocks may have overlapping periods. When calculating the mean square value, the mean square unit 122 may use signal values ​​within a period (step period) excluding a period that overlaps with the immediately following gating block (overlap period), and may ignore signal values ​​within other periods. The mean square unit 122 outputs the calculated mean square z to the weighting unit 124. The weighting unit 124 calculates a weighted mean square value by multiplying the mean square value z input from the mean square value unit 122 by a weighting coefficient G set in advance for that channel. The weighting unit 124 outputs the calculated weighted mean square value to the addition unit 126. For binaural signals, the weighting coefficient is set to an equal value between channels. The adder 126 calculates an inter-channel weighted average value by summing the weighted root mean square values ​​for each channel input from the weighting unit 124. The adder 126 outputs the calculated inter-channel weighted average value to the decibel converter 128.

[0024] The decibel converter 128 converts the inter-channel weighted average value input from the adder 126 into a decibel value. The decibel value is obtained by multiplying the common logarithm with base 10 by 10. The decibel converter 128 outputs the converted decibel value to the gate processor 130. The gate processing unit 130 calculates a gating block loudness value by adding a predetermined coefficient to the decibel value input from the decibel conversion unit 128. The gate processing unit 130 determines whether the determined gating block loudness value is greater than a predetermined gating threshold value.

[0025] The gate processing unit 130 adopts the gating block loudness value for each gating block whose gating block loudness value exceeds the gating threshold during the period T, and discards the other gating block loudness values. The gate processing unit 130 calculates the average of the adopted gating block loudness values ​​as the gated loudness value for the period T. Since loudness values ​​during very short silent periods that the listener does not perceive as silence are discarded, the loudness of the sound is quantified in accordance with the auditory sensation. The gate processing unit 130 outputs the calculated gated loudness value to the outside of the loudness measuring device 10.

[0026] The filter control unit 140 controls whether or not to perform the first filtering process, depending on whether or not the acoustic signal input to the characteristic filtering processing unit 110 is a binaural signal. If the acoustic signal input to the characteristic filtering processing unit 110 is a binaural signal, the filter control unit 140 generates a first control signal indicating that the first filtering process is required, and outputs the generated first control signal to the switch unit 112. If the acoustic signal input to the characteristic filtering processing unit 110 is not a binaural signal, the filter control unit 140 generates a first control signal indicating that the first filtering process is not required, and outputs the generated first control signal to the switch unit 112. An example of determining whether or not the acoustic signal input to the characteristic filtering processing unit 110 is a binaural signal will be described later.

[0027] The filter control unit 140 may be configured to always perform the second filter processing. This is because the transfer characteristics indicated by the HRIRs or HRTFs (hereinafter collectively referred to as "HRIRs, etc.") used to generate binaural signals are purely physical characteristics and do not include psychological characteristics such as human hearing. However, the gain in the low frequencies of HRIRs, etc. obtained by measurement or synthesis may be extremely low depending on the conditions at the time of acquisition. For example, this is because auditory correction may be applied to one or a combination of the speakers used to play back sound, the microphones used to collect sound, and the mixer used for editing. In such a case, performing the second filter processing reduces the contribution to the loudness value in the low frequencies.

[0028] Therefore, filter control unit 140 may control whether or not to perform second filtering depending on whether or not the gain in a low band whose frequency is lower than a predetermined reference frequency is equal to or greater than a predetermined reference gain. If the gain in the low band is equal to or greater than the reference gain, filter control unit 140 generates a second control signal indicating that second filtering is required and outputs the generated second control signal to switch unit 116. If the gain in the low band is not equal to or greater than the reference gain, filter control unit 140 generates a second control signal indicating that second filtering is not required and outputs the generated second control signal to switch unit 116. An example of determining whether the gain in the low band is equal to or greater than the reference gain will be described later.

[0029] (Example of binaural signal determination) Next, an example of determining whether or not the acoustic signal input to the characteristic filter processing unit 110 is a binaural signal will be described. For example, when a binaural signal is specified as the acoustic signal to be measured by a user operation, the filter control unit 140 determines not to perform the first filtering process, and when another acoustic signal is specified, to perform the first filtering process. Information specified by the user's processing can be transmitted by an operation signal input from an operation input unit (not shown). The operation input unit includes a component that receives a user operation and generates an operation signal in accordance with the received operation. Such a component may be a dedicated component such as a button, lever, or knob, or may be a general-purpose component such as a touch sensor or mouse.

[0030] Furthermore, when the characteristic filter processing unit 110 is notified of the input of two-channel acoustic signals from the binaural signal generation unit 150 (described later), the characteristic filter processing unit 110 may determine not to execute the first filter processing. When detecting the input of two-channel acoustic signals from the binaural signal generation unit 150, the characteristic filter processing unit 110 can notify the input of two-channel acoustic signals by outputting signal input information indicating the input to the filter control unit 140. If the binaural signal generation unit 150 is detachable from other components of the loudness measurement device 10, the filter control unit 140 may be provided with a detection unit (not shown) that detects the attachment of the binaural signal generation unit 150. When wearing information indicating the attachment of the binaural signal generation unit 150 is input from the detection unit, the filter control unit 140 may determine not to execute the first filter processing, and when wearing information indicating the non-attachment of the binaural signal generation unit 150 is input, the filter control unit 140 may determine to execute the first filter processing. The detection unit may include a contact sensor that detects physical contact with the binaural signal generation unit 150, or a continuity sensor that detects electrical connection with the binaural signal generation unit 150.

[0031] Furthermore, the filter control unit 140 may analyze the two-channel audio signal input to the characteristic filter processing unit 110 and determine whether the input audio signal is a binaural signal. The filter control unit 140 determines that the input audio signal is a binaural signal when, for low-frequency components whose frequencies are lower than a predetermined reference frequency, the correlation coefficient between channels is equal to or greater than a predetermined correlation coefficient threshold, and when, among a plurality of frequency bands in the high frequency range whose frequencies are equal to or greater than the reference frequency, the variance of the correlation coefficient between channels for each frequency band is greater than a predetermined variance threshold; otherwise, the filter control unit 140 can determine that the input audio signal is not a binaural signal. This allows the filter control unit 140 to determine whether the input audio signal is a binaural signal by capturing the characteristics that the correlation between channels is high in the low range and the variance of the correlation between channels is significant in the high range.

[0032] (Example of gain determination in low frequencies) Next, an example of determining whether the gain in the low frequency range is equal to or greater than a reference gain will be described. The filter control unit 140 reads out the HRIRs and the like used to generate the binaural signal from the binaural signal generation unit 150. The filter control unit 140 analyzes the transfer characteristics indicated in the read-out HRIRs and the like, and can determine whether the gain of the transfer characteristics in the low frequency range, which is equal to or less than a reference frequency, is equal to or greater than a predetermined reference gain. The filter control unit 140 analyzes the acoustic signal input to the characteristic filter processing unit 110, and can determine whether the gain of the transfer characteristic in the low frequency range is equal to or greater than a predetermined reference gain based on whether the ratio of the power in the low frequency range, where the frequency is equal to or less than the reference frequency, to the power at the reference frequency is equal to or greater than a predetermined reference power ratio.

[0033] (Binaural signal generation section) Next, another example of the functional configuration of the loudness measuring device 10 according to this embodiment will be described. As illustrated in FIG. 2, the loudness measuring device 10 may further include a binaural signal generation unit 150. The binaural signal generation unit 150 generates, from the sound source signal s, a left ear signal x1 for presenting sound to the left ear of the listener, and a right ear signal x2 for presenting sound to the right ear. The generated pair of left ear signal and right ear signal is output as a binaural signal to the characteristic filter processing unit 110. The binaural signal generation unit 150 can generate a left ear signal by adding a transfer characteristic from the sound source direction, which is the direction in which the sound source is installed, to the sound source signal, and can generate a right ear signal by adding a transfer characteristic from the sound source direction to the right ear.

[0034] The binaural signal generating unit 150 includes a storage unit 152 and a convolution operation unit 156 . HRIR data is stored in advance in the storage unit 152. The HRIR data is data indicating an HRIR that indicates the transfer characteristics from a sound source to the left ear for each sound source direction, and an HRIR that indicates the transfer characteristics from the sound source to the right ear. The HRIRs from a sound source placed in the sound source direction to each of the left and right ears are read out by the convolution calculation unit 156.

[0035] The convolution calculation unit 156 reads out, from the storage unit 152, HRIRs from a sound source direction corresponding to a target direction for binaural reproduction to each of the left and right ears. The convolution calculation unit 156 performs a convolution calculation on the input sound source signal using the HRIR up to the left ear to generate a signal for the left ear, and performs a convolution calculation on the sound source signal using the HRIR up to the right ear to generate a signal for the right ear. The convolution calculation unit 156 outputs the generated left ear signal and right ear signal to characteristic filter processing units 110-1 and 110-2, respectively.

[0036] Note that a fixed target direction may be preset in the convolution calculation unit 156, or target direction information indicating a target direction that may change over time may be input (rendering). The convolution calculation unit 156 identifies the target direction indicated in the input target direction information and reads out the HRIR corresponding to the identified target direction from the storage unit 152. For example, information having a format conforming to the ADM specified in Non-Patent Document 2 may be used as the target direction information, or information indicated by an operation signal input from an operation input unit (not shown) may be used. The operation input unit may be built into or connected to the loudness measuring device 10, or may be built into or connected to an external device (e.g., a mixing console) separate from the loudness measuring device 10.

[0037] 2 illustrates an example of a functional configuration for generating two-channel binaural signals from one system of sound source signals corresponding to one sound source, but the present invention is not limited to this. The binaural signal generation unit 150 may generate a sound source-specific left ear signal and a sound source-specific right ear signal for each of multiple sound sources. The binaural signal generation unit 150 may further add (mix) the sound source-specific left ear signals between the sound sources to generate an output left ear signal, and add (mix) the sound source-specific right ear signals between the sound sources to generate an output right ear signal. In this case, the target direction may be set for each system of sound source signals corresponding to each individual sound source.

[0038] FIG. 3 illustrates a functional configuration for generating two-channel binaural signals from N (N is a predetermined integer equal to or greater than 2) systems of sound source signals. The convolution calculation unit 156 includes N sets of HRIR convolution units 157-1L to 157-NR and two adders 158-1L and 158-1R. The HRIR convolution units 157-nL and 157-nR (n is an integer between 1 and N) each perform a convolution calculation of the HRIR from the nth sound source direction to the left ear on a sound source signal related to the nth sound source to generate a sound-source-specific left-ear signal for the nth sound source, and perform a convolution calculation of the HRIR from the nth sound source direction to the right ear to generate a sound-source-specific right-ear signal for the nth sound source. The HRIR convolution units 157-nL and 157-nR output the generated sound-source-specific left-ear signal to the adder 158-1L and output the generated sound-source-specific right-ear signal to the adder 158-1R. The addition unit 158-1L adds the sound source-specific left ear signals input to itself between the N sound sources to generate an output left ear signal, and outputs the generated output left ear signal to the characteristic filter processing unit 110-1 as an acoustic signal x1. The addition unit 158-1R adds the sound source-specific right ear signals input to itself between N sound sources to generate an output right ear signal, and outputs the generated output right ear signal to the characteristic filter processing unit 110-2 as an acoustic signal x2.

[0039] In addition to rendering, the convolution unit 156 may also be applied to multi-channel sound reproduction. To apply it to multi-channel sound reproduction, first, HRIRs from each sound source position defined in the playback method to be used to each of the left and right ears of a listener at the listening position are acquired in advance. Each sound source position corresponds to the installation position of a playback sound source for each channel. Then, for each of the left and right ears, a convolution operation is performed on the sound source signal using the HRIR from each playback sound source to the ear, and the resulting sound source-specific signals are mixed between N sound sources to obtain signals for each ear. For example, to apply it to 22.2ch sound, HRIRs from the sound source positions where each of the 24 (=N) channel playback sound sources defined in Non-Patent Document 1 is installed to each of the left and right ears of a listener at a predetermined listening position are measured in advance. The HRIR convolution unit 157 convolves the sound source signal with the HRIR measured for each of the left and right ears for each channel to generate a sound source-specific left ear signal and a sound source-specific right ear signal. However, the HRIRs acquired in advance are set in the HRIR convolution units 157 for the corresponding channels and ears. The 24-channel acoustic signals are supplied to a pair of left and right HRIR convolution units 157 for each channel. Therefore, sounds similar to those coming from 24 sound sources are presented to each of the listener's left and right ears as sounds based on the generated binaural signals.

[0040] The applicable multi-channel sound reproduction method is not limited to 22.2ch sound, but may also be 5.1ch sound, 7.1ch sound, etc. For example, in 5.1ch sound and 7.1ch sound, the positional relationships between six and eight sound source positions and the listening position are specified, respectively. HRIRs can be set in the HRIR convolution unit 157 according to these positional relationships. In this embodiment, instead of performing a convolution operation of the HRIR with the acoustic signal, the acoustic signal may be converted into the frequency domain to obtain a conversion coefficient, which may be multiplied by the HRTF corresponding to the HRIR, and the multiplied value obtained by the multiplication may be converted into the time domain.

[0041] As described above, the loudness measurement device 10 according to this embodiment includes a characteristic filter processing unit 110 that receives audio signals of multiple channels and performs first filter processing to add predetermined spatial acoustic characteristics to the audio signals of each channel, and a second filter processing unit 118 that performs second filter processing to add predetermined auditory characteristics to the audio signals input from the first filter processing unit 114; a loudness calculation unit 120 that determines a loudness value based on the intensity obtained by weighting the power of the audio signals input from the characteristic filter processing unit 110 across channels; and a filter control unit 140 that determines whether or not to perform the first filter processing depending on whether the audio signals of the multiple channels are binaural signals. According to this configuration, the first filter processing is not executed when a binaural signal is input to the characteristic filter processing unit 110. Therefore, the spatial acoustic characteristics included in the binaural signal and the spatial acoustic characteristics resulting from the first filter processing do not overlap, and a loudness value that conforms to the auditory sensation can be obtained. Furthermore, since the first filter processing is executed when a binaural signal is not input to the characteristic filter processing unit 110, the requirements for existing loudness measurements can be satisfied.

[0042] The binaural signal may include a left ear signal for sound presented to the left ear and a right ear signal for sound presented to the right ear, and may include a binaural signal generation unit 150 that generates a signal for the left ear by adding a transfer characteristic from the sound source to the left ear to the sound source signal, and generates a signal for the right ear by adding a transfer characteristic from the sound source to the right ear to the sound source signal. According to this configuration, a binaural signal is generated based on a sound source signal, and a loudness value for the generated binaural signal is obtained, thereby making it possible to efficiently edit or produce content that includes the binaural signal based on the obtained loudness value.

[0043] The filter control section 140 may determine whether or not to execute the second filtering process depending on whether or not the gain of the spatial acoustic characteristics at low frequencies equal to or lower than a predetermined reference frequency is equal to or higher than a predetermined reference gain. According to this configuration, the second filtering process is performed when the gain of the transfer characteristic in the low frequency range is high, so that a loudness value that matches the auditory sense can be obtained in consideration of the low contribution to the auditory sense in the low frequency range.

[0044] The loudness measuring device 10 may be realized as a measuring device dedicated to measuring loudness values, or may be realized as a device whose main function is not to measure loudness values, such as an information terminal device such as a personal computer or a tablet terminal device. The loudness measuring device 10 may also be realized as part of equipment (e.g., a mixing console) related to the production, editing, and distribution (including broadcasting) of various types of content.

[0045] Note that some or all of the loudness measurement device 10 described above, for example, any one or a combination of the characteristic filter processing unit 110, the loudness calculation unit 120, the filter control unit 140, and the binaural signal generation unit 150, may be implemented by a computer. In this case, a program for implementing each control function may be recorded on a computer-readable recording medium, and the program may be loaded and executed by a computer system. Note that the term "computer system" used here refers to a computer system built into the loudness measurement device 10, including hardware such as an operating system (OS) and peripheral devices. Furthermore, the term "computer-readable recording medium" refers to portable media such as flexible disks, optical magnetic disks, read-only memories (ROMs), and CD-ROMs, as well as storage devices such as hard disks built into a computer system. Furthermore, the term "computer-readable recording medium" may also include a medium that dynamically stores a program for a short period of time, such as a communication line when transmitting a program via a network such as the Internet or a communication line such as a telephone line, or a medium that stores a program for a certain period of time, such as volatile memory within a computer system that serves as a server or client in such a case. Furthermore, the above program may be one that realizes part of the functions described above, or may be one that can realize the functions described above in combination with a program already recorded in the computer system.

[0046] Furthermore, part or all of the loudness measuring device 10 in the above-described embodiment may be realized as an integrated circuit such as an LSI (Large Scale Integration). Each functional block of the loudness measuring device 10 may be individually implemented as a processor, or part or all of the functional blocks may be integrated into a processor. Furthermore, the integrated circuit implementation method is not limited to LSI, and may be implemented using a dedicated circuit or a general-purpose processor. Furthermore, if an integrated circuit implementation technology that can replace LSI emerges due to advances in semiconductor technology, an integrated circuit based on that technology may be used.

[0047] One embodiment of the present invention has been described in detail above with reference to the drawings, but the specific configuration is not limited to that described above, and various design changes and the like are possible within the scope that does not deviate from the gist of the present invention. [Explanation of symbols]

[0048] 10... loudness measuring device, 110 (110-1, 110-2)... characteristic filter processing section, 112 (112-1, 112-2), 116 (116-1, 116-2)... switch section, 114 (114-1, 114-2)... first filter processing section, 118 (118-1, 118-2)... second filter processing section, 120... loudness calculation section, 122 (122-1, 122-2)... characteristic filter processing section, 116 (116-1, 116-2)... switch section, 116 (116-1, 116-2)... first filter processing section, 116 (116-1, 116-2)... second filter processing section, 120... loudness calculation section, 122 (122-1, 122-2)... 2-2)...mean square unit, 124 (124-1, 124-2)...weighting unit, 126...adder, 128...decibel converter, 130...gate processor, 140...filter controller, 150...binaural signal generator, 152...storage unit, 156...convolution unit, 157 (157-1L to 157-NR)...HRIR convolution unit, 158 (158-1L, 158-1R)...adder

Claims

1. A multi-channel acoustic signal is input, a first filter processing unit that performs a first filter process on the acoustic signal of each channel using a filter that simulates an acoustic effect caused by a human head; a characteristic filter processing unit including a second filter processing unit that executes second filter processing to add predetermined auditory characteristics to the acoustic signal input from the first filter processing unit; a loudness calculation unit that determines a loudness value based on an intensity obtained by weighted averaging the power of the acoustic signal input from the characteristic filter processing unit across channels; a filter control unit that determines not to perform the first filtering process when the acoustic signals of the multiple channels are binaural signals, and to perform the first filtering process when the acoustic signals of the multiple channels are not binaural signals. Loudness measuring device.

2. A loudness measurement device comprising a binaural signal generation unit, the binaural signals include a left-ear signal for sound presented to the left ear and a right-ear signal for sound presented to the right ear, The binaural signal generation unit generating a signal for the left ear by adding a transfer characteristic from the sound source to the left ear to the sound source signal; generating a signal for the right ear by adding a transfer characteristic from the sound source to the right ear to the sound source signal; 2. The loudness measuring device according to claim 1.

3. The filter control unit Whether or not to perform the second filtering process is determined depending on whether or not the gain of the transfer characteristic at a low frequency equal to or lower than a predetermined reference frequency is equal to or higher than a predetermined reference gain.

3. The loudness measuring device according to claim 2.

4. A program for causing a computer to function as the loudness measuring device according to any one of claims 1 to 3.

Images