Acoustic signal processing device

Abstract

To provide an "acoustic signal processing device" capable of reducing occurrence of crosstalk without impairing stability of a sound reproduction system using a stereo speaker. The solution is to filter an input stereo acoustic signal by a filter processing unit (2) and output the filtered acoustic signal from a stereo speaker (3). By setting a filter transfer function of the filter processing unit (2) to a function having no denominator and only a numerator, and the function including a sign-inverted transfer function of a transfer function representing spatial acoustic characteristics of a crosstalk component of sound reaching a listening point (4) from the stereo speaker (3), it is possible to eliminate instability of the filter transfer function diverging infinitely large with the denominator becoming zero, and to output sound from the stereo speaker (3) in which the crosstalk component is eliminated in a transfer path from the stereo speaker (3) to the listening point (4).

Description

Technical Field

[0001] This disclosure relates to audio signal processing devices. Background Technology

[0002] The spatial image of a sound source perceived by a person is called a "sound image." Furthermore, the spatial location of the perceived sound image (the direction and distance from which the sound is heard) is called "sound image localization." Previously, sound image localization devices were known to assign a sound image to an input audio signal, positioning it at a predetermined spatial location. By appropriately controlling sound image localization, it is possible to reproduce more realistic stereo sound or to give the sound reaching the listener's ears a sense of direction. For example, it is possible to assign a sound image positioned as if the sound were heard from a specific direction.

[0003] Patent Document 1 discloses a sound image localization device that can clearly identify the location of a sound source in sound reproduction using headphones or over-ear headphones. However, when using headphones or over-ear headphones for sound image localization, if the listener moves their head in the direction of the sound when they perceive it, there is a problem that the headphones or over-ear headphones move along with the head, and the direction of the sound also moves at the same time.

[0004] Patent Document 2 discloses a sound image localization device that, when mixing 5ch (five-channel) audio sources into 2ch (two-channel) sound, uses a model head transfer function from the rear to both ears to add back localization, and performs crosstalk cancellation correction to ensure the effectiveness of this added back localization effect. In the case of a reproduction system using speakers instead of headphones or headsets, crosstalk problems exist, such as sound intended for the right ear reaching the left ear and sound intended for the left ear reaching the right ear. In the sound image localization device described in Patent Document 2, crosstalk is eliminated when mixing 5ch surround sound into 2ch stereo sound.

[0005] One method to address this crosstalk problem is a technique called transaural recording. A transaural system uses a simulated human head to record sound in both ears and then reproduces the recorded sound using stereo speakers instead of headphones. In a transaural system, a filter is used to calculate the transfer function obtained by inversely operating on the signal reproducing from the 2ch speakers as it travels from the speakers to the listening points (both ear positions). At the listening points in the reproduced sound field, the system attempts to reproduce the original binaural recording (sound without crosstalk) from the original sound field.

[0006] Figure 5 It is a diagram schematically representing the structure of the auditory transmission system. In Figure 5 In the original sound field, the sound emitted from the sound source S is subject to the spatial transfer function H.r The signal reaches the right ear due to the influence of the spatial transfer function H. l The sound reaches the left ear due to the influence of the sound source S. On the other hand, when the recorded sound from the sound source S is emitted as is from the speakers 101 and 102 (left channel and right channel) in the reproduced sound field, it is affected by the spatial transfer function H. R,r and H L,r The affected sound reaches the right ear, therefore the sound reaching the right ear will not be the same as the sound emitted by Rch's speaker 101, but will be different. Similarly, due to the spatial transfer function H... R,l and H L,l The affected sound reaches the left ear, so the sound reaching the left ear will not be the same as the sound emitted by Lch's loudspeaker 102, but will be different. In contrast, the auditory transmission system is an acoustic filter (a filter with characteristics opposite to spatial transmission characteristics) that has a transfer function G that eliminates the effects of spatial transmission characteristics, and filters the listening signal obtained in the original sound field, so that the listening sounds l and r in the reproduced sound field become the same listening sounds l and r as in the original sound field.

[0007] The transfer function G of the acoustic filter used in the auditory transmission system is expressed as shown in Equation 0. In Equation 0, the transfer function {H} L,l (ω)H R,r (ω)-H L,r (ω)H R,l (ω)} -1 Represents the transfer function {H} L,l (ω)H R,r (ω) -H L,r (ω)H R,l The inverse transfer function of (ω)} (also known as the inverse system or inverse filter). As shown in Equation 0, regarding the transfer function G set for the acoustic filter in the auditory transmission system, the transfer function is included in the denominator. Therefore, depending on the condition of the reproduced sound field, the calculation result of the transfer function G becomes unstable, and there is a problem that the stable operation of the system cannot be guaranteed.

[0008] Existing technical documents: Patent documents: Patent Document 1: Japanese Patent Application Publication No. 2002-209300 Patent Document 2: Japanese Patent Application Publication No. 2007-110206 Summary of the Invention

[0009] The problem that the invention aims to solve: This disclosure was made to address such a problem, with the aim of reducing crosstalk in sound reproduction systems using stereo speakers without compromising system stability.

[0010] Methods used to solve problems: To address the aforementioned issues, the audio signal processing apparatus of this disclosure is configured to include a filter processing unit that filters an input stereo audio signal, and outputs the filtered audio signal from a stereo speaker. The filter transfer function of the filter processing unit is a function with only a numerator and no denominator, and it is a function that includes a sign-inverted transfer function, which is obtained by inverting the sign of a transfer function representing the spatial acoustic characteristics of crosstalk components of sound arriving at the left and right listening points from the stereo speaker.

[0011] Invention effects: According to the present disclosure configured as described above, by using a function with only the numerator and no denominator as the filter transfer function of the filter processing unit, the instability of the filter transfer function diverging to infinity when the denominator becomes zero can be eliminated. Furthermore, since a sign-inverted transfer function containing crosstalk components is used as the filter transfer function of the filter processing unit, sound output from the stereo speaker can have crosstalk components eliminated in the transmission path from the stereo speaker to the left and right listening points. Therefore, in a sound reproduction system using a stereo speaker, the occurrence of crosstalk can be reduced without compromising system stability. Attached Figure Description

[0012] Figure 1A This is a diagram illustrating an example of the configuration of the audio signal processing device according to this embodiment.

[0013] Figure 1B This diagram illustrates other configuration examples of the audio signal processing apparatus according to this embodiment.

[0014] Figure 1C This diagram illustrates other configuration examples of the audio signal processing apparatus according to this embodiment.

[0015] Figure 2 This is a diagram illustrating a configuration example of the audio signal processing device of the first modified example.

[0016] Figure 3 This is a diagram illustrating the configuration example of the audio signal processing device in the second variation.

[0017] Figure 4 This is a diagram illustrating the configuration example of the audio signal processing device in the third modified example.

[0018] Figure 5It is a diagram that schematically represents the structure of the auditory transmission system.

[0019] Explanation of reference numerals in the attached figures: 1…Head transfer function processing unit, 2…Filter processing unit, 2'…First adjustment filter unit, 2''…Second adjustment filter unit, 3…Stereo speaker, 4…Listening point, 5A, 5B…Spatial transfer analog processing unit. Detailed Implementation

[0020] Hereinafter, one embodiment of the present disclosure will be described based on the accompanying drawings. Figure 1A As shown in (a), the audio signal processing device of this embodiment is configured to include a head transfer function processing unit 1 and a filter processing unit 2.

[0021] Head transfer function processing unit 1 processes the head transfer functions H for the left and right sides. l (z), H r The filter coefficients of (z) filter the audio signal from the sound source to generate stereo audio signals l(z) and r(z). Additionally, in the following explanation, it may sometimes be expressed as "filtering by the transfer function," but this is an abbreviation meaning "filtering by setting the filter coefficients used to install the transfer function."

[0022] The Head Related Transfer Function (HRTF) is a function that represents the transmission characteristics of sound waves as they travel through the head, auricle, and other parts of the ear to the entrance of the listener's external auditory canal. It is a function that depends on physical characteristics such as the shape of the listener's head and auricle.

[0023] As an example, sound output from speakers positioned around the simulated head is collected at the entrance of the external auditory canal, and the head transfer function H is measured by performing known signal processing on the collected sound signals. l (z), H r (z). The head transfer function processing unit 1 is composed of a filter in which the head transfer function H is measured. l (z), H r (z) is set as a filter coefficient. This can be achieved through the set head transfer function H. l (z), H r (z) is used to perform the desired sound image localization. For example, it can make the sound reaching the left and right listening points 4 (the left and right ears of the listener or the left and right microphones) directional.

[0024] The filter processing unit 2 filters the stereo sound signals l(z) and r(z) input from the head transfer function processing unit 1 using a predetermined transfer function, and outputs the filtered stereo sound signals from the stereo speakers 3 (left and right speakers 3L and 3R). The filter processing unit 2 is composed of an FIR (finite impulse response) filter with filter coefficients set for installing the predetermined transfer function.

[0025] like Figure 1A As shown in (a), the filter transfer function of the filter processing unit 2 is as follows: the transfer function H represents the spatial acoustic characteristics from the sound output from the right loudspeaker 3R to the right listening point 4R. Rr The transfer function H represents the spatial acoustic characteristics from the sound output from the right-hand loudspeaker 3R to the left-hand listening point 4L. Rl The transfer function after sign reversal -H Rl The function H represents the spatial acoustic characteristics from the sound output from the left loudspeaker 3L to the right listening point 4R. Lr The transfer function after sign reversal -H Lr And the transfer function H, which represents the spatial acoustic characteristics from the sound output from the left loudspeaker 3L to the left listening point 4L. Ll Two transfer functions H Rr H Ll The two transfer functions H represent the spatial acoustic characteristics related to the direct components of the sound reaching the left and right listening points 4L and 4R. Rl H Lr This indicates the spatial acoustic characteristics related to the crosstalk component in the sound reaching the left and right listening points 4L and 4R.

[0026] The four filter transfer functions H of the filter installed in the filter processing unit 2 Rr -H Rl -H Lr H Ll All with Figure 5 The transfer function G shown is different; it is a function with only a numerator and no denominator. Additionally, two of the four transfer functions, -H... Rl -H Lr It is a sign-inverted transfer function, which is the transfer function representing the spatial acoustic characteristics of crosstalk components, and the crosstalk components are the sounds that arrive at the left and right listening points 4R and 4L on the opposite sides from the left and right loudspeakers 3L and 3R, respectively.

[0027] Filter processing unit 2 uses these four transfer functions H to perform the operation. Rr -H Rl -H Lr H LlThe filter's multipliers and adders perform multiplication and addition operations on the stereo audio signals l(z) and r(z) input from the head transfer function processing unit 1 to filter the signal, and output the filtered stereo audio signal from the stereo speaker 3. Equation 1 below represents the multiplication and addition operation performed in the filter processing unit 2.

[0028] Figure 1A (b) shows the relationship with Figure 1A The mathematically equivalent configuration of the audio signal processing device shown in (a) is as follows: That is, the stereo audio signals l(z) and r(z) output from the head transfer function processing unit 1 are subjected to... Figure 1A The influence of the acoustic characteristics modeled by the transfer function shown in (b) propagates to the listening point 4 (the left and right listening points 4L and 4R). As shown in equation (2) below, the four transfer functions H installed in the filter processing unit 2 are used to propagate the sound characteristics. Rr -H Rl -H Lr H Ll The four spatial transfer functions H represent the spatial acoustic characteristics from the left and right loudspeakers 3L, 3R to the left and right listening points 4L, 4R. Rr H Rl H Lr H Ll Multiplication and addition operations between them are used to calculate Figure 1A The transfer function is shown in (b).

[0029] like Figure 1A As shown in (b), in the transmission path from the head transfer function processing unit 1 to the listening point 4 via the filter processing unit 2 and the stereo speaker 3, crosstalk components are eliminated and only direct components are obtained. Furthermore, the transfer functions associated with the audio signal l(z) of Lch and the audio signal r(z) of Rch are equal to each other.

[0030] Furthermore, in order to enable adjustments to bring the input-output characteristics represented by Equation 2 above closer to the desired input-output characteristics, it is also possible to increase the number of input-output components with... Figure 1B The filter coefficients W shown are as follows lL (z), W rR (z) is the first adjustment filter section 2'. With this configuration, for example, it is possible to make the characteristics of the transmission system that remain after the elimination of crosstalk components approach the desired characteristics.

[0031] In addition, Figure 1AThe four spatial transfer functions H used in the filter processing unit 2 of (a) Rr H Rl H Lr H Ll Ideally, spatial transfer functions (SJFs) should be obtained individually for each person wearing an earpiece microphone. However, in reality, most SJFs are represented by those obtained using a simulator or by measurements taken by a specific person wearing an earpiece microphone. In this case, due to the relationship between the individual's SJF H and the actual SJF H, the SJF H is significantly different from the actual SJF H. Rr H Rl H Lr H Ll Slightly different, therefore, to correct this deviation, it is also possible to utilize something with... Figure 1C The filter coefficients W, which specifically include crosstalk components, are thus added. lL W rL W lR W rR The second adjustment filter unit 2'' is a two-input, two-output filter matrix system used to adjust both the direct component and crosstalk component of the sound arriving at the listening point 4. This is done to convert the sound into a signal that is eliminated by the filter processing unit 2. Figure 1A The system with the crosstalk component shown in (b) if the following is adopted Figure 1C The configuration shown allows for easy setting / adjustment of the final mixed crosstalk components using only the second adjustment filter section 2''.

[0032] As described above, in the audio signal processing apparatus of this embodiment, a function with only a numerator and no denominator is used as the filter transfer function of the filter processing unit 2. Therefore, the instability of the filter transfer function diverging to infinity when the denominator becomes zero can be eliminated. Furthermore, since a sign-inverted transfer function containing crosstalk components is used as the filter transfer function of the filter processing unit 2, sound output from the stereo speaker 3 with crosstalk components eliminated in the transmission path from the stereo speaker 3 to the listening point 4 can be output. Thus, in a sound reproduction system using the stereo speaker 3, the occurrence of crosstalk can be reduced without compromising system stability. Additionally, by adding a first adjustment filter unit 2', the input-output characteristics of the two systems obtained by reducing crosstalk can be adjusted to the desired input-output characteristics. Furthermore, if a second adjustment filter unit 2'' is used, the setting / adjustment of the final mixed crosstalk components can be easily performed.

[0033] The following uses Figures 2-4 A variation of the audio signal processing apparatus of this embodiment will be described. Figures 2-4The variations shown all further incorporate the spatial transmission simulation processing units 5A and 5B, detailed below. The spatial transmission simulation processing units 5A and 5B filter the stereo sound signal filtered by the filter processing unit 2 using an analog transfer function that simulates a predetermined spatial sound characteristic from the stereo speaker 3 to the left and right listening points 4L and 4R. Furthermore, the stereo sound signal filtered by the filter processing unit 2 and the spatial transmission simulation processing units 5A and 5B is output from the stereo speaker 3.

[0034] Figure 2 This is a diagram illustrating the configuration example of the audio signal processing device in the first modified example. (See diagram for example.) Figure 2 As shown in (a), the audio signal processing apparatus of the first modified example includes a spatial transfer analog processing unit 5A after the filter processing unit 2. The analog transfer function of the filter installed in the spatial transfer analog processing unit 5A is a function with only a numerator and no denominator, and is a function that does not include the transfer function representing the spatial acoustic characteristics of crosstalk components.

[0035] Specifically, the analog transfer function of the spatial transfer simulation processing unit 5A is a transfer function H that simulates the spatial acoustic characteristics directly related to the distance from the left loudspeaker 3L to the left listening point 4L. Ll And the transfer function H, which simulates the spatial acoustic characteristics related to the direct components from the right-hand loudspeaker 3R to the right listening point 4R. Rr .

[0036] The spatial transfer simulation processing unit 5A utilizes the transfer function H, which is related to the direct component on the left. Ll The audio signal of Lch, after being filtered by filter processing unit 2, is filtered, and the transfer function H, which is related to the direct component on the right, is used. Rr The audio signal of Rch, which has been filtered by the filter processing unit 2, is filtered. Furthermore, the stereo audio signal filtered by the spatial transmission analog processing unit 5A is output from the stereo speaker 3.

[0037] Figure 2 The audio signal processing device shown in (a) can be approximately represented as Figure 2 The system shown in (b) is such that [lr] and [Y] l Y r The relationship can be expressed as follows (Equation 3).

[0038] According to the audio signal processing apparatus of this first modified example, the spatial transmission analog processing unit 5A filters the sound using an analog transfer function that simulates the spatial acoustic characteristics of the direct component of the sound originally required at the listening point 4. Then, during the period from when the sound actually output from the stereo speaker 3 propagates through space to reach the listening point 4, the spatial acoustic characteristics of the direct component are further applied. That is, when the sound reaches the listening point 4 under the condition that the influence of the direct component is applied twice, the required spatial acoustic characteristics of the direct component are emphasized and reproduced at the ear. Thus, the sound of the direct component can be reproduced more richly while suppressing crosstalk components. Furthermore, since the head transfer function is further convolved in the head transfer function processing unit 1, the listener can hear the sound at the listening point 4 with a greater emphasis on the characteristics of directionality and distance.

[0039] Figure 3 This is a diagram illustrating the configuration example of the audio signal processing device in the second modified example. (See diagram for example.) Figure 3 As shown in (a), the second modified audio signal processing apparatus includes a spatial transfer analog processing unit 5B after the filter processing unit 2. The analog transfer function of the filter installed in the spatial transfer analog processing unit 5B is a function with only a numerator and no denominator, and includes a transfer function representing the spatial acoustic characteristics of crosstalk components.

[0040] Specifically, the analog transfer function of the spatial transfer simulation processing unit 5B includes two transfer functions H that simulate the spatial acoustic characteristics related to the direct components of the sound from the stereo speaker 3 to the left and right listening points 4L and 4R. Rr H Ll ; and two transfer functions H that simulate the spatial acoustic characteristics related to crosstalk components in the sound arriving at the left and right listening points 4L and 4R from the stereo speaker 3. Rl H Lr .

[0041] The space transfer simulation processing unit 5B uses four transfer functions H. Rr H Rl H Lr H Ll The multiplier and adder of the filter perform multiplication and addition operations to filter the stereo signal filtered by the filter processing unit 2. Furthermore, the stereo signal filtered by the spatial transmission analog processing unit 5B is output from the stereo speaker 3.

[0042] Figure 3 The audio signal processing device shown in (a) can be approximately represented as Figure 3 The system shown in (b) is such that [lr] and [Y] l Y rThe relationship can be expressed as shown in Equation 4 below. Furthermore, the relationship expressed by Equation 4... Figure 3 The system of (b) and in Figure 1C Set as W lL =H Ll W lR =H Rl W rL =H lR W rR =H Rr The system is equivalent.

[0043] In this second modified example of the audio signal processing apparatus, the spatial transmission analog processing unit 5B performs filtering using an analog transfer function that simulates the spatial acoustic characteristics of sound including both direct and crosstalk components. Then, during the period from when the sound actually output from the stereo speaker 3 travels through space to reach the listening point 4, the same spatial acoustic characteristics are further applied. That is, with crosstalk components suppressed by the processing of the filter processing unit 2 and the spatial transmission analog processing unit 5B, the sound is input to the stereo speaker 3 and reaches the listening point 4 in such a way that the influence of spatial acoustic characteristics is applied only in the final stage. In other words, the influence of the direct component can be implemented in two ways, and the state of the crosstalk component can be adjusted as needed, thereby richer reproduction of the sense of distance and expansion of the sound.

[0044] Figure 4 This is a diagram illustrating the configuration example of the audio signal processing device in the third modified example. (See diagram for example.) Figure 4 As shown, the audio signal processing device of the third modification will... Figure 3 The combination of the filter processing unit 2 and the spatial transmission analog processing unit 5B shown herein constitutes a multi-stage cascaded connection. Furthermore, in Figure 4 The illustration of listening point 4 is omitted in the text.

[0045] According to the audio signal processing device of the third modified example configured in this way, the first and second stage filter processing units 2 -1 2 -2 The processing suppresses crosstalk components, and the simulation processing unit 5B transmits spatial information through the first and second stages. -1 5B -2 The processing applies the same acoustic characteristics to the stereo signal as those from the stereo speaker 3 to the listening point 4. Therefore, with the spatial acoustic characteristics from the stereo speaker 3 to the listening point 4 further enhanced by multiple applications, the sound reaching the listening point 4 can more richly reproduce the sense of distance and expansion of the sound.

[0046] In addition, Figure 4The diagram shows a configuration example in which the filter processing unit 2 and the spatial transmission analog processing unit 5B are cascaded in two stages, but the cascaded connection can also be three stages or more.

[0047] Alternatively, the first and second variations can be applied in combination, or the first and third variations can be applied in combination. For example, the space transfer simulation processing unit 5B (or the space transfer simulation processing unit 5B) can also be used in this way. -2 The analog processing unit 5A is connected between the stereo speaker 3 and the spatial transmission unit 5A.

[0048] In the above embodiments and the first to third modifications, a configuration including a head transfer function processing unit 1 was described, but this is not a necessary configuration. By including the head transfer function processing unit 1, the desired sound image can be located in the sound reaching the listener's ear at the listening point 4. However, even without the head transfer function processing unit 1, the effect of reducing crosstalk without compromising the stability of the system can be achieved.

[0049] Furthermore, the above embodiments are merely examples illustrating specific implementations of this disclosure, and the technical scope of this disclosure should not be interpreted as limiting. That is, this disclosure can be implemented in various forms without departing from its spirit or main features.

Claims

1. An audio signal processing device, characterized in that, The system includes a filter processing unit that filters the input stereo signal using filter coefficients with a predetermined transfer function. The audio signal processing device is configured to output a stereo audio signal filtered by the filter processing unit from a stereo speaker. The filter transfer function of the filter processing unit is set to be a function with only numerator and no denominator, and the function includes a sign-inverted transfer function, which is a function that inverts the sign of the transfer function that represents the spatial acoustic characteristics of the crosstalk component of the sound from the stereo speaker to the left and right listening points.

2. The audio signal processing device according to claim 1, characterized in that, It also includes a spatial transfer simulation processing unit, which filters the stereo signal filtered by the filter processing unit using filter coefficients equipped with an analog transfer function. The analog transfer function simulates a specified spatial acoustic characteristic from the stereo speaker to the left and right listening points. The simulated transfer function is set as a function with only numerator and no denominator. The audio signal processing device is configured to output a stereo audio signal filtered by the filter processing unit and the spatial transmission analog processing unit from the stereo speaker.

3. The audio signal processing device according to claim 2, characterized in that, The simulated transfer function is a function that does not include a transfer function representing the spatial acoustic characteristics of the crosstalk component. It is a transfer function that simulates the spatial acoustic characteristics related to the direct component from the left loudspeaker to the left listening point, and a transfer function that simulates the spatial acoustic characteristics related to the direct component from the right loudspeaker to the right listening point.

4. The audio signal processing device according to claim 2, characterized in that, The simulated transfer function is a function that includes a transfer function representing the spatial acoustic characteristics of the crosstalk component, and is a transfer function that simulates the spatial acoustic characteristics from the stereo loudspeaker to the left and right listening points.

5. The audio signal processing device according to claim 4, characterized in that, The audio signal processing device is formed by cascading multiple stages of the filter processing unit and the spatial transmission analog processing unit.

6. The audio signal processing apparatus according to any one of claims 1 to 5, characterized in that, It also includes a head transfer function processing unit, which uses filter coefficients with a head transfer function installed to filter the audio signal to generate a stereo audio signal. The filter processing unit receives the stereo sound signal output from the head transfer function processing unit.

7. The audio signal processing device according to claim 1, characterized in that, The filter processing unit is further provided with a first adjustment filter unit in front of it. The first adjustment filter unit is used to adjust the input-output characteristics from the input stage input to the filter processing unit to the listening point.

8. The audio signal processing device according to claim 1, characterized in that, The pre-stage of the filter processing unit also includes a second adjustment filter unit, which is used to adjust both the direct component and the crosstalk component of the sound arriving at the listening point.

Images