Audio processor

Abstract

There is provided an audio processor (16) suitable for adapting audible output of an output device (18). The audio processor (16) being configured to: receive an audio input signal; receive an identified frequency response of an output device configuration across a frequency range; generate equalisation filtering based on the identified frequency response; and provide an audio output signal to the output device (18), the audio output signal including the audio input signal with the equalisation filtering applied. A corresponding method is also provided.

Description

FIELD OF THE INVENTION

[0001] The present disclosure generally relates to audio processing, typically concerning adapting the audible output, such as from headphones, to take account of the effect of the audio production configuration on the audible output.BACKGROUND

[0002] Significant effort is applied when building audio production equipment to keep the form of the output as close a replication of the source as possible. This keeps the output "transparent" or undistorted rather than adding "colour" or distortion.

[0003] Keeping an output or signal transparent allows effects to be applied to a signal that is as close to "true" or undistorted as possible rather than needing to try to address detrimental consequences of the production equipment with effects. However, achieving this is complex and is typically a balancing act between resources, whether that be cost or power or some other limiting factor, and optimised output.

[0004] Additionally, a single entity rarely has control of an entire audio production chain from audio creation to audio output. This means equipment outside of an entity's control is also able to affect the signal, the effects of which may not be able to be addressed by the entity's equipment.

[0005] Audio output is provided through headphones or by loudspeakers. These are connected to an output of an item of production equipment, such as an amplifier, audio processor, synthesiser, Musical Instrument Digital Interface (MIDI) device, mixer, (record, cassette, CD, etc.) deck, computer, or some other audio production equipment.

[0006] The audible audio output from an output device (e.g. a headphone set or loudspeaker) is coloured by the configuration of the output device. For wired connections, the audible audio output is also coloured by the connection to the production equipment from which the output device receives an audio signal. For headphones, this is typically a connection to a headphone amplifier, which is often, but not always, part of another piece of production equipment.

[0007] For some headphone sets the configuration can be changed, such as by changing the earpad types. This affects the audible output from these headphone sets due to the change in materials. Additionally, the interaction of impedance of a headphone set and the impedance of the output to which the headphone set is connected can cause the frequency response of the headphone set to be coloured.

[0008] Headphone sets are available with different impedances. Picking a headphone set with a different impedance from the impedance of the output to which it is connected assists with reducing the colouring caused by this interaction. Indeed, some have proposed a solution of ensuring the headphone impedance is at least five times or even eight times the impedance of the output to which the headphone set is connected to provide a more transparent sound. However, this requires access to multiple sets of headphones or causes undesired output quality if a user only has low impedance headphones available to connect to a high impedance output. Additionally, this cannot take account of other aspects of a headphone set configuration altering the audible output, such as the type of earpad used, which is thus most clearly addressed by changing earpad type. However, this can affect user comfort or have other detrimental effects.

[0009] Accordingly, there is a need to provide a more transparent audible output from headphones regardless of overall configuration of an audio production chain.SUMMARY OF INVENTION

[0010] According to a first aspect, there is provided an audio processor suitable for modulating audible output of an (audio) output device, the audio processor being configured to: receive an audio input signal; receive an identified frequency response of an output device configuration across a frequency range; generate equalisation filtering based on the identified frequency response; and provide an audio output signal to the output device, the audio output signal including the audio input signal with the equalisation filtering applied.

[0011] Rather than reconfiguring the output device, or changing the output device, this allows the same output device to be used by adapting the signal to take account of the frequency response of the output device.

[0012] Typically, output devices, such as loudspeakers and headphones are selected based on the reported quality of the audio output relative to the price bracket a user is applying. This means it is usually unclear what other production equipment the output device is to be used with, or at least that the user will wish to be able to use the output device with a variety of production equipment. The instinct is therefore for adjustments to be made to the hardware of the output device to tailor its frequency response. This is because it allows such a tailored frequency response to be applicable to all production equipment the output device could be used with and not subject to variable quality electronics or calibration of electronics. However, we have found that providing an audio processor that is able to account for the frequency response of the output device through filtering provides a more reliable and versatile functionality, and thus offers an improved experience for the user.

[0013] The terminology "audible output" is intended to mean the output able to be heard by a user, such as the audio output received by a user. This is intended to be more than simply the audio output of a device. This is because audio that is output by a device, or at least, what is received by a user, is subject to interference or distortion by the environment between where the audio is output and where it is received. It is possible to measure audible output of a device as well as to identify intended audio output.

[0014] The output device is able to be any form of audio output device, including a loudspeaker. Typically, however, the output device is a headphone set, such as a pair of headphones or a single headphone. The terms "headphone set", "pair of headphones", "headphone pair", "single headphone", "headphones" or "headphone" are used interchangeably herein unless specifically stated otherwise.

[0015] The audio processor being configured to receive the identified frequency response may be an electronic frequency response or an audio frequency response. Further, the identified frequency response may be a transfer function. The audio processor may be further configured to identify the frequency response of the output device. This is in order for, or part of, the identified frequency response being received. Alternatively, the receiving of the identified frequency response may be due to the identified frequency response being provided to the audio processor, such as by a user, by identification in a library or from an external input, or by an algorithm configured to calculate a frequency response based on input parameters, such as output device frequency response or impedance and output impedance of the audio input device.

[0016] The frequency range may be a frequency range typically audible to (normal hearing) humans, such as 20 (or 15) Hertz (Hz) to 20,000 Hz, may be (an extended range such as) 10 Hz to 40,000 Hz or may be between 5 Hz and 100,000 Hz.

[0017] The identified frequency response of the output device may be any frequency response of the output device, or may include any attribute, property or influencing factor on the frequency response. Typically, the identified frequency response of the output device configuration includes an impedance of the output device configuration across the frequency range. Typically, the output impedance is stable at least over the normal audible frequency range of humans (i.e. 15 or 20 Hz to 20,000 Hz as set out above), and usually well beyond that range too. This is because the output stage is typically resistive.

[0018] While, when provided by a supplier, an impedance of an (audio) output device is typically quoted as a single value, such as 30 Ohms (Ω). Typically, that is the impedance at a particular frequency (usually a nominal number that represents the lowest point of a midband of an impedance curve of the output device). As such, while impedance is often reported as a single value in relation to audio equipment, a relevant impedance of any output device or any piece of audio equipment is typically a range of values, or varies, across a range of frequencies. This is because headphones and loudspeakers have a complex impedance consisting of resistive and reactive components.

[0019] This allows the equalisation to compensate for the impedance of the output device across the frequency range. The impedance of an audio output device (such as a headphone set or loudspeaker) is commonly not the same across the frequency range. Instead, the impedance response varies across the frequency range. This distorts an audio signal, and, due to the variation, is a variable distortion. This is often undesirable, especially when a user is looking for a transparent output. Accordingly, being able to compensate for the impedance variation significantly reduces the distortion of the output signal, providing an enhanced user experience and greater accuracy in the reproduction of the audio input signal. Further, this can reduce the amount of further processing a user may want to apply, thereby limiting processing power and equipment requirements.

[0020] The audio input signal may be received from an audio input device. This may result in the audio processor being further configured to receive an identified frequency response of the audio input device, and the equalisation filtering generation being further based on the identified frequency response of the audio input device, and thus based on the identified frequency response of the audio output device and the identified frequency response of the audio input device. Typically, but not always, this is when the identified frequency response of the output device includes an impedance of the output device configuration across the frequency range.

[0021] The combination of the equalisation being based on the frequency response, for example the impedance frequency response, of each of the audio input device and (audio) output device allows the equalisation to take account of the interaction between these frequency responses. While simply taking account of the frequency response of the audio output device will provide improved transparency of the audio output signal, there will be interaction between the two frequency responses. This means taking account of the interaction will further optimise the reduction in distortion.

[0022] It is noted that various forms of frequency response are referred to throughout. With the introduction above of two forms of frequency response, it is noted that various frequency responses may include a (mostly) resistive output impedance, such as of an audio input device, which does not significantly change with frequency, i.e. is a very "flat" frequency response; a frequency varying output device (e.g. headphone) impedance, which can be referred to as an "impedance curve"; and a resulting audible frequency response a user listens to or hears.

[0023] The source (e.g. the audio input device) output frequency response does not change significantly with frequency, hence being described as "flat". Additionally, output device designers attempt to provide output devices, such as headphones and loudspeakers, that have a user perceived flat frequency response. However, to achieve that, the frequency response is tailored to the specifics of the model of the device or the specific device itself, so is typically not flat. Typically, without intervention in accordance with aspects described herein, it is this last frequency response alters due to the impedance interaction between the audio input device and the output device.

[0024] The audio input device is intended to be an audio interface. The audio interface may, in some circumstances, also, or alternatively, be an audio generation device, such as a drum machine or musical instrument digital interface (MIDI) hardware, but this is not required. As such, the audio input device may be arranged in use to provide processing of an audio signal it receives from an audio generation device before providing that signal as an audio input signal to the audio processor of the first aspect. In some cases, the audio processor and audio input device may be part of, or modules of, the same piece of hardware.

[0025] Unlike audio output devices, audio input devices, when an audio interface or a piece of audio production equipment rather than purely an audio generation device that does not include a processor, the output impedance may be a constant value. By this it is intended to mean that the value is constant across the frequency range. There are, however, circumstances in which this output impedance may vary across the frequency range.

[0026] As a result of the output impedance of the audio input device and the impedance of the audio output device across the frequency range interacting, there is distortion of the audio signal output provided to the audio output device, or, in other words, the electrical signal passed to the audio output device. As such, if not taken into account, the identified frequency response of the audio input device, a default audio input device frequency response, such as a default output impedance, would likely be used to provide one part that causes the distortion.

[0027] Typically though, as set out above, the audio input device frequency response is provided. When the audio input device frequency response is provided as an output impedance of the audio input device across the frequency range, the audio processor may be further configured to calculate an audio signal distortion across the frequency range of an interaction of output impedance of the audio input device across the frequency range and the impedance of the output device configuration across the frequency range, the generated equalisation filtering being a reproduction of the calculated distortion as a negative.

[0028] This allows an imitation of the distortion to be generated to try to at least partially cancel out the distortion. The distortion will typically be an increase or decrease in signal strength or equivalent audio output volume from the true audio signal, and thus is an increase or decrease from zero. The equalisation filtering is intended to provide the opposite distorting effect on the audio signal to the distortion caused by the interaction of the output impedance and impedance of the output device. This replication thus diminishes the distortion.

[0029] The identified frequency response of the audio input device may be identified by the (specific type or model of) audio input device being provided as an input, such as by a user or as data from a source, including, but not limited to, user input or by data transmission from the audio input device or some other device connected to the audio processor and / or the audio input device. In these circumstances, the audio processor may include a lookup table or be able to access a lookup table, the lookup table including a correspondence between a plurality of audio input devices and frequency response. The audio process may then read a respective frequency response from the lookup table, receiving the respective frequency response as the identified frequency response. Typically, however, the identified frequency response of the audio input device may be received as an input provided to the audio processor. This reduces the storage capacity requirements and / or connectivity requirements of the audio processor, and limits data access and processing latency since it limits database access and read operations.

[0030] The input (of the identified frequency response) may be a value selected from a range or may be a section of the range selected from the range. The selection may be a user selection. This allows the user to select the appropriate frequency response for the audio input device being used, either by selecting a specific value (such as an impedance value), for example using a dial, slider or numerical input field, or by selecting a section of a range from a larger overall range (such as a sub-range of impedance values in an overall impedance range), for example using a drop-down menu, radio buttons, toggles, sliders or a (numerical) input field. The use of a specific value allows accurate selection of the frequency response and the use of a section of a range allows an approximate selection to be made. Since, if the extent of the section is suitably restricted, the differences in the frequency response at the maximum and minimum values in each section are not significant enough for it to affect the user experience. For example, each section may be a 20 Ω impedance range or a 10 Ω impedance range with the overall range extending from 0 Ω to 80 Ω or 100 Ω, such as 0 Ω to 20 Ω, 20 Ω to 40 Ω, 40 Ω to 60 Ω, 60 Ω to 80 Ω and 80 Ω to 100 Ω. The extent or resolution of each range (i.e. difference between the lower limit and upper limit of each range) is able to be based on what a user is able to perceive. For example, a resolution or range extent of 0.1 dB will provide audio differences that are imperceptible to a user. This means that having a set of ranges that cause an output to only vary by 0.1dB between adjacent ranges of impedance settings is less likely to be worthwhile. However, if a difference in output has a 0.5 dB to 1.0 dB resolution, that would be more worthwhile, since this resolution is on the threshold of audibility.

[0031] Additionally or alternatively to the identified frequency response of the output device configuration including an impedance of the output device configuration across the frequency range, the identified frequency response of the output device configuration may include an audible frequency response across the frequency range of the output device configuration. This allows the audio output signal to be modified to take account of factors that affect the audible frequencies emitted by the output device, such as materials used and placement of components, especially when these have been altered from a standard, default or designed configuration.

[0032] The audible frequency response across the frequency range may be a (pre)measured audible frequency response, but could be a calculated audible frequency response. The audible frequency response being measured allows a non-theoretical frequency response to be used. Since the audible frequency response will be affected by specific installation arrangements and manufacturing tolerances, using a theoretical audible frequency response is likely to be less accurate than a measured audible frequency response.

[0033] The output device may be any form of audio output device, such as a loudspeaker. Typically, however, the output device may be a headphone set and the output device configuration is the headphone set with one of a plurality of earpad types. In this case, the measured audible frequency response may be a measured audible frequency response of the headphone set with an identified earpad type of the plurality of earpad types. This allows the effect of earpad type on the audible audio output to be accounted for.

[0034] The earpad type may be identified by a user providing the earpad type, such as from a discrete list of earpad types or by providing data regarding earpad type. For a list of earpad types, the audio processor may have access to details of the audible frequency response for each earpad type, which is accessed to receive the identified frequency response.

[0035] The identified frequency response of the output device may be identified by the (specific type or model of) output device being provided as an input, such as by a user or as data from a source, including, but not limited to, user input or by data transmission from some other device connected to the audio processor or by identification or selection of a frequency response or impedance value. As such, typically, the identified frequency response of the output device configuration is received based on input provided to the audio processor.

[0036] Since the output device typically has a frequency response that varies across the frequency range, providing a single value for the frequency response may not be representative of the actual frequency response. This is because even if two output devices, such as devices from different manufacturers, have the same reported frequency response if reported as a single value, the response of those output devices across the frequency range may vary from each other. Thus, typically, the audio processor is further configured to read a stored database entry selected based on input provided to the audio processor, the stored database entry being the identified frequency response of the output device configuration.

[0037] The audio processor may have a database of, or have access to a database of, one or more frequency responses from which it may be able to read entries. This allows a simple lookup operation to be conducted instead of needing to calculate the frequency response. While this would be possible by measuring an amount of current drawn by the output device across a range of voltage frequencies, this is a more complex operation that can be desirable, but, at minimum, requires calibration. Accordingly, reading a stored database entry and using that as the identified frequency response is a less complex operation and provides a simplified operation. When a user provides input of the output device this allows the user choice of the selection avoiding an automated system mis-identifying the output device or frequency response.

[0038] The audio processor may be configured to generate the equalisation filtering (by one of): calculating a filter as a reproduction of a negative of the identified frequency response(s); or identifying a filter in a lookup table from the identified frequency response(s). Calculating the filter to apply allows dynamic creation of the filter, allowing lower storage capacity or access to lower storage capacity for the audio processor. On the other hand, identifying the filter using a lookup table allows lower processing power.

[0039] The filter may be one (or more) of: a biquad filter set; an infinite impulse response filter; a finite impulse response filter; or a parametric equalisation filter and (corresponding) centre value. These allow accurate reproduction of the frequency response(s) in order to counteract the distortion produced by the frequency response(s) by emulating the frequency response.

[0040] According to a second aspect, there is provided a method of modulating audible output of an output device, the method comprising: receiving an audio input signal; receiving an identified frequency response of an output device configuration across a frequency range; generating equalisation filtering based on the identified frequency response; and providing an audio output signal to the output device, the audio output signal including the audio input signal with the equalisation filtering applied.

[0041] The method according to the second aspect may apply any one or more of the functions set out above or may implement features of the audio processor according to the first aspect is able to perform or is configured to provide.

[0042] According to a third aspect, there is provided an audio processor suitable for modulating audible output of an output device, the audio processor being configured to: receive an audio input signal from an audio input device; receive an identified frequency response of an output device configuration across a frequency range and an identified frequency response of an audio input device across the frequency range; generate equalisation filtering based on the identified frequency response of the output device configuration and the identified frequency response of the audio input device; and provide an audio output signal to the output device, the audio output signal including the audio input signal with the equalisation filtering applied.

[0043] Typically, the identified frequency response of the output device configuration may include an impedance of the output device configuration across the frequency range and the identified frequency response of the audio input device may include an output impedance of the audio input device across the frequency range. In these circumstances, the audio processor may be further configured to calculate an audio signal distortion across the frequency range of an interaction of output impedance of the audio input device across the frequency range and the impedance of the output device configuration across the frequency range, the generated equalisation filtering being a reproduction of the calculated distortion as a negative.

[0044] According to a fourth aspect, there is provided a method of modulating audible output of an output device, the method comprising: receiving an audio input signal from an audio input device; receiving an identified frequency response of an output device configuration across a frequency range and an identified frequency response of an audio input device across the frequency range; generating equalisation filtering based on the identified frequency response of the output device configuration and the identified frequency response of the audio input device; and providing an audio output signal to the output device, the audio output signal including the audio input signal with the equalisation filtering applied.

[0045] According to a fifth aspect, there is provided an audio processor suitable for modulating audible output of an output device, the audio processor being configured to: receive an audio input signal; receive an identified audible frequency response of an output device configuration across a frequency range; generate equalisation filtering based on the identified frequency response; and provide an audio output signal to the output device, the audio output signal including the audio input signal with the equalisation filtering applied.

[0046] Typically, the output device may be a headphone set and the output device configuration may be the headphone set with one of a plurality of earpad types. The audible frequency response across the frequency range may be a measured audible frequency response. In such a case, the measured audible frequency response may be a measured audible frequency response of the headphone set with an identified earpad type of the plurality of earpad types.

[0047] According to a sixth aspect, there is provided a method of modulating audible output of an output device, the method comprising: receiving an audio input signal; receiving an identified audible frequency response of an output device configuration across a frequency range; generating equalisation filtering based on the identified audible frequency response; and providing an audio output signal to the output device, the audio output signal including the audio input signal with the equalisation filtering applied.

[0048] According to a seventh aspect, there is provided a computer program comprising instructions which, when executed, cause an apparatus to perform the method according to any one of the second, fourth and / or sixth aspects.

[0049] The computer program may be in the form of a plug-in, i.e. a software component that adds a specific functionality to another program. The other program may be a digital audio workstation (DAW). Alternatively, the computer program may be a standalone program that operates separately from other programs.

[0050] According to an eighth aspect, there is provided a non-transitory computer-readable medium comprising the computer program according to the seventh aspect.BRIEF DESCRIPTION OF DRAWINGS

[0051] Example audio processors and methods are described in detail below with reference to the accompanying drawings, in which: Figure 1 shows an example schematic arrangement including an example audio processor; Figure 2 shows an example output device impedance curve; Figure 3 shows an example distortion curve and equalisation filtering; Figure 4 shows a first example interface; Figure 5 shows a second example interface; Figure 6 shows an example process; and Figure 7 shows a schematic diagram of an example non-transitory computer-readable medium. DETAILED DESCRIPTION

[0052] An arrangement implementing an example audio processor is generally illustrated at 1 in Figure 1. The arrangement shown in Figure 1 includes example audio generation devices in the form of a microphone 10 and guitar 12. In other examples, the audio generation devices may take another form, such as a different instrument or audio generation device. These include synthesisers, streamed audio reproduced via an audio player, and record, cassette or CD decks.

[0053] The audio generation devices are connected to an audio input device 14 by standard connection means, such as audio cables. In various examples, the audio input device is an audio interface, such as a Focusrite ®< Scarlett ®< 2i2 ®< . Other audio input devices include other audio production equipment, such as an amplifier, MIDI device and mixer.

[0054] In some examples, the audio input device 14 is also an audio generation device, or is able to access stored (or itself stores) audio input signals. In such examples, separate audio generation devices need not be included.

[0055] The audio input device 14 is connected to an audio processor 16 and an audio output device 18. In the example shown in Figure 1, the audio output device is a headphone set. In other examples, the audio output device can be another form of audio output device, such as a loudspeaker.

[0056] The audio processor 16 is represented in the example shown in Figure 1 by a computer, but is able to be any form of audio processor. In various examples, however, the audio processor is a DAW, a plug-in in a DAW or a standalone (software) application. This allows the audio processor to be formed by a processor (of the computer) and to provide a user interface via the display of the computer. In other examples, however, the audio input device may also be the audio processor, or the audio processor may be provided by another device or computer program.

[0057] In the example shown in Figure 1, the audio input device 14 has an input connection to the audio processor provided by an input 142, such as a MIDI connection or USB connection, and an output connection from the audio processor provided by an output 162, such as a MIDI connection or USB connection. In some examples, the input and the output are provided by the same connection.

[0058] The audio processor 16 also has an audio output connected, typically by a cable, to the audio output device 18. An audio output signal is provided to the audio output device by this audio output.

[0059] Whether the audio output device 18 is a headphone set or loudspeaker, this will have an impedance. For headphones, the reported impedance is typically between about 8 Ω and 600 Ω. Headphones with an impedance between about 8 Ω and 50 Ω are considered "low impedance" headphones and headphones with an impedance above about 50 Ω are considered "high impedance" headphones.

[0060] Loudspeakers typically have a reported impedance between about 2 Ω and 16 Ω, with most speakers having an impedance of between about 4 Ω and 8 Ω.

[0061] Regardless of the reported impedance of a headphone set or loudspeaker, the impedance is not typically constant or level across the range of operable frequencies, including over the range of human audible frequencies. Instead, as shown in the example impedance curve 180 in Figure 2, the impedance varies across a frequency range.

[0062] In Figure 2, a plot of frequency (in Hertz, Hz) against impedance (in Ohms, Ω) is shown with a base 10 log scale for the frequency. This plot shows an impedance curve 180 for an example headphone set. Between 1 Hz and 10 Hz there is a slow, slightly upwardly curved increase in the impedance; between 10 Hz and about 80 Hz there is a curved, rapid, increase in the impedance, increasing about 12 times the amount the curve increased between 1 Hz and 10 Hz. From this peak, the impedance then drops down to a similar level as it is at 10 Hz by a frequency between about 1000 Hz and 1100 Hz. The impedance then increases in a shallow curve, increasing in gradient towards 100000 Hz. This plot is representative (if potentially simplified) of an impedance curve for a headphone set.

[0063] Loudspeakers also have well-known and / or recognisable typical impedance curves. The specifics of the impedance curve for a headphone set or loudspeaker is dependent on a number of factors and may vary significantly from the example plot shown in Figure 2.

[0064] It is not only the audio output device 18 that has an impedance, however. The audio output of the audio input device also has an impedance. This is commonly referred to as "output impedance". This corresponds to the amount of current able to be provided by the respective audio output. Devices considered to have a low output impedance usually, but not always, provide a high amount of current and devices considered to have a high output impedance usually, but not always, provide a low amount of current. What is, in fact, able to be provided is, however, typically, determined by a combination of output impedance, amplifier and / or power supply unit of / incorporated within the device. Regarding the effect of impedance, this is due to the impedance being able to be considered in some circumstances as a restriction on current flow.

[0065] When reported, output impedances of typical audio input devices can have a very large range from impedances of less than 1 Ω to impedances in excess of two factors of ten larger. While output impedance can vary across the frequency range, the output impedance is typically approximately constant, and level, across the frequency range shown in Figure 2 for the impedance curve.

[0066] Without an output device connected to its audio output, audio input devices are typically designed to have a flat frequency response. An example of a flat frequency response can be seen in the distortion plot of Figure 3 at 200 as a line lying along the horizontal axis origin (i.e. 0).

[0067] Figure 3 shows a plot of frequency against distortion (typically measured in decibels, dB). In addition to the example flat frequency response 200, this shows two other curves.

[0068] As noted above, the flat frequency response 200 is the frequency response of the audio input device when no output device is connected to the audio output. This is commonly referred to as "unloaded" due to the output device presenting a load to the audio input device when connected. Once connected, the frequency response changes. This changes due to the frequency response, including the impedance, of the output device across the frequency range and the interaction of that with the frequency response, including the output impedance, of the audio input device (and specifically, in various examples, the audio output of the audio input device to which the output device is connected).

[0069] Figure 3 shows an example distortion curve 210 of an audio signal across the frequency range when an output device is connected to an audio input device. The frequency response shape of the output device broadly corresponds to the example response shown in Figure 2, although may be shifted relative to the frequency range compared to the curve shown in Figure 2.

[0070] The example distortion curve 201 of Figure 3 shows a bell curve-like shape between 1 Hz and about 900 Hz with a peak between 10 Hz and 20 Hz. The curve then has a positive gradient as the frequency increases towards 100000 Hz.

[0071] When the distortion is at a level of around 1 dB or above, a listener is able to detect the distortion in the audible audio output. For trained sound engineers, the distortion can be identified at lower levels, but a 1 dB distortion level would generally be accepted as being audible. This is considered undesirable since it causes the volume of different sections of the frequency range to be different from each other when the intention is that the volume is constant across the frequency range (i.e. the frequency range audible to humans).

[0072] A further cause of distortion in a headphone set is distortion caused by the audible frequency response by the earpad material. There are several types of headphone, which include "in-ear", "over-ear" or "on-ear". For at least the latter two types, there are also "open back" and "closed back" (i.e. have a back or outward-facing portion that is, at least partially, open to the air or has a solid back) varieties.

[0073] Additionally, headphone sets can have earpads that can be changed. In on-ear and over-ear headphone sets, these comprise a cushion in a ring or (generally) circular shape. The size and shape of the cushion varies from model to model. The cushion provides a separation between a user's ear and the speaker in the headphone. This thereby forms a cavity that, in use, is defined by the cushion, headphone structure, speaker and the user's ear. The size, shape and composition of this cavity along with the corresponding characteristics of the headphone structure, speaker and user's ear all affect the audible sound across the frequency range of an audio output of the headphone set.

[0074] For in-ear headphone sets, there is still able to be a cavity defined by the portion of the headphone that sits within the user's ear, speaker or speaker grill and the user's ear. There may still be cushioning provided for such headphone sets with a form of earpad being provided by the portion of the headphone that sits within the user's ear, and which may be cushioning in some headphone sets.

[0075] The available variations in earpads allow different types of earpad to be used. For example, the earpad material may have an outer material or outer layer that is real or artificial leather, (soft) cloth or other fabrics, foam, plastic, silicone or other materials. As well as different outer materials or layers, earpads may be vented, such as by having one or more holes or perforations through which air is able to pass or shaped in a particular manner. The internal material or layers of an earpad may also vary, such as by foam pore size or compressibility, shape and / or composition. An earpad may also include one or more foils or other materials, or exclude one or more foils or other materials from the internal material(s). The earpad may further include a felt, foam or mesh layer between the cavity and the speaker within the headphone body.

[0076] There are several reasons to change the earpad type, such as comfort, number of expected users, and frequency and length of use. The earpad type (for example any of the combinations of material above) provides part of the output device configuration. This means that when the earpad type is changed, the output device configuration changes.

[0077] Due to the differences in composition and configuration of earpad types, this alters the effect of the earpad type on the cavity provided, and thus on the audible audio output from a headphone set. In view of this, headphone sets are designed for only a single earpad type or a theoretical average of some or all of the available earpad types for the headphone set. Should the user change the earpad type from the earpad type it is purchased with, or for which it is designed, this will thus alter the audible audio output of the headphone set. This alteration can be significant and can alter across the frequency range.

[0078] Sometimes this alteration can be desirable or sought out. For example, the Sennheiser ®< HD 490 Pro has two earpad sets, one for production of a velvet-like material, and one for mixing of a denim-like material. The production variety provides increased bass volumes, providing "warmth", whereas the mixing variety provides a flatter response up to about 1,000 Hz. While this alteration in the audible audio output is identified as beneficial in this situation, it is often undesirable.

[0079] It is possible to measure the difference in audible output produced by different earpad types. For an example headphone set, this measurement is achieved by using an ear coupler.

[0080] An ear coupler typically provides one flat plate per headphone. Each flat plate may have a rubber-like material and potentially an ear-like shape. This also includes a canal intended to replicate the human ear canal. One or more microphones are located within this canal.

[0081] In use a headphone is placed against the flat plate of the ear coupler in the orientation in which is mounted to the ear, sometimes held on with a jig to provide consistent placing when testing. Audio output is then played through the headphone that provides an output across the frequency range. This test is carried out with the headphone set for each earpad type (i.e. by having one earpad type attached to the headphone set for each test). By recording the output, it is then possible to identify the difference in audible audio output signal between different earpad types, and, in any case, the audible audio output across the frequency range for each earpad type.

[0082] In some examples, the difference in audible audio output signal of a headphone set with one earpad type compared to another or to a default earpad type may be similar to the distortion curve 210 shown in Figure 3. However, typically, this is different. Regardless, distortion still occurs.

[0083] As noted above, whatever the source of the distortion, this may be sought out or may be limited by changing output device configuration (whether that be due to a change in the device itself or just a change in the configuration). We have found that it is, instead, possible to address this distortion with "EQ" or equalisation, also referred to as equalisation filtering. This is due to a recognition of what contributes to the distortion.

[0084] We have recognised that the interaction causing the impedance related distortion is effectively due to a change in electrical dampening. What is meant by this is that, in effect, the output impedance of the audio input device and the impedance of the output device over the frequency range form a voltage divider. This divider provides different dampening behaviour depending on the frequency.

[0085] As noted above, the audio input device output impedance is (approximately) the same across the relevant frequency range, but can be different from one model of device to another; and the output device impedance varies with frequency with the characteristics of this variance changing from model to model. In the illustrative example of a voltage divider, it is possible to consider each of the audio input device and output device as two resistors (each representing the impedance of the respective device) connected in series. The resistor that represents the audio input device is a fixed resistance due to the consistent impedance across the relevant frequency range. In turn, the resistor that represents the output device is a variable resistance due to the impedance varying across the relevant frequency range. This means that when a voltage is applied across the opposing ends of these resistors (i.e. when the devices are in use) the voltage measurable between the mid-point between the resistors and one of the ends varies as the resistor representing the output device varies its resistance, but with the variation amount affected by both resistors.

[0086] This varying voltage is analogous to the distortion curve discussed above. This means that the characteristics of the distortion curve are affected by the output impedance of the audio input device as well as the impedance of the output device.

[0087] By providing EQ that is the negative of the distortion curve, an undistorted output is able to be provided. Since the audio output signal when passing from the audio input device to the output device is simply an electronic signal with a varying voltage, the provided EQ adjusts that voltage by application of a negative of the distortion curve, an undistorted output is able to be provided. In other words, this is because the negative curve, such as the example dashed line curve 220 shown in Figure 3, cancels the distortion that would otherwise be present.

[0088] To mimic or replicate a negative version of the distortion curve of the audible audio output signal, frequency response of the output device (such as due to its impedance) or the interaction of the frequency response of each of the output device and the audio input device (such as due to the impedance of the output device and the output impedance of the audio input device), EQ filtering is generated in use. We have identified this is possible to achieve when the contributions to the distortion are known, or at least the characteristics of the contributions. This is because, when the varying frequency response of the output device is known, it is possible to model the resulting distortion due to the contribution of a fixed frequency response. Once the varying frequency response and an example fixed frequency response are known, the distortion curve for that combination of parameters can be calculated. However, this means knowledge derived from multiple devices along the audio production chain is used. This has not previously been considered since, instead, focus has been on refining an optimising individual devices in the chain, and due to parties typically being limited to supplying products in one portion of the audio production chain only.

[0089] With this recognition, we have identified that it is possible for filtering to reproduce, as closely as possible, a negative version of the distortion curve. This can be achieved by use of at least one form of filter. These include the following forms of filter: a biquad filter set; an infinite impulse response (IIR) filter; a finite impulse response (FIR) filter; a parametric / semi-parametric / quasi-parametric equalisation filter with a centre value for the filter; low cut filter; high cut filter; low shelf filter; high shelf filter; bell curve filter; band pass filter; high pass filter; low pass filter; notch filter; bass tone filter; treble tone filter; and peak filter.

[0090] In use, the filter(s) or filter combination can be selected as appropriate based on the frequency response across the frequency range of the output device configuration in terms of, in a number of examples, the impedance and / or audible audio output. This can also be selected, in some examples, based on the frequency response across the frequency range of the audio input device, such as the impedance of the audio input device. Additionally, or alternatively, this can be selected, in some examples, due to its phase characteristics to equalise also any phase distortions. Typically, an FIR filter would be used for this. In various examples, this is able to be based, at least partially, on the interaction of the impedance across the frequency range of the output device configuration and the impedance across the frequency range of the audio input device.

[0091] The filter(s) or filter combination can be generated by an algorithm configured to create filters to meet the relevant criteria. Alternatively, an algorithm may select filters from a library, database or look-up table. In other examples, the appropriate filters to use for one or more output device configurations and, in some examples, one or more audio input device or audio input device frequency responses are stored in a library, database or look-up table. The relevant entry is able to be read from the library, database or look-up table on receipt of the relevant frequency responses, for example, by a library, database, or look-up table query.

[0092] To be able to apply appropriate filtering, identification of the characteristics for the various scenarios able to be modelled is needed. In terms of how this information is received by the audio processor, in some examples, the frequency response of the output device and, when relevant, the audio input device is identified automatically. In other examples, this information is identified or provided by a user or is based on information provided by the user. Examples of this are shown in Figure 4 and Figure 5.

[0093] Figure 4 shows a user interface 400 of the audio processor, in this case as (part of) an example plug-in for a DAW. This user interface has three user inputs in the example shown. In other examples, there can be more user inputs or less user inputs, and / or the user inputs are able to be provided in other forms.

[0094] The inputs shown in example of Figure 4 are in two categories. The categories are impedance and earpad type (shown in Figure 4 as "Earpad Material").

[0095] The impedance category has two inputs in the example shown in Figure 4. These are an example first drop-down menu 410 for headphone (identified in Figure 4 as "HP") model and an example second drop-down menu 420 for impedance. The earpad type category has one input in the example shown in Figure 4. This is a toggle switch 430 for earpad material.

[0096] In the example shown in Figure 4, the impedance category inputs allow a user to identify a headphone set model being used as an output device and an impedance of an audio input device. The first drop-down menu provides a list of options for the user to choose from in various examples. Similarly, in some examples, the second drop-down menu allows the user to either select a model (and optionally make) of an audio input device or an output impedance of the audio input device.

[0097] In various examples, the selections made in the first drop-down menu 410 and the second drop-down menu 420, identify a location in a library, database or look-up table. This location, in a number of examples, stores a predetermined frequency response of the interaction, across the frequency range, of the impedance of the headphone model and output impedance or audio input model (and optionally make). In some examples, the relevant one or more filters are then identified that produce a negative of the above frequency response, such as by generating the one or more filters or reading details of the one or more filters from a further library, database or look-up table.

[0098] In other examples, the location in the library database or look-up table provides details of the relevant one or more filters for each combination of selections from the first drop-down menu 410 and the second drop-down menu 420. This means that user selection from those menus provides the appropriate one or more filters to apply to produce a negative of a predetermined frequency response of the interaction, across the frequency range, of the impedance of the headphone model and output impedance or audio input device.

[0099] In some examples, the selection in the first drop-down menu identifies a location in a library, database or look-up table of a predetermined impedance of the selected headphone model across the frequency range and, together with the selection from the second drop-down, the interaction, across the frequency range, of the impedance of the headphone model and output impedance of the audio input device is calculated by the audio processor.

[0100] In various circumstances, one or more loudspeaker models may be included additionally or as an alternative to headphone models in the example first drop-down menu 410.

[0101] The predetermined impedance and / or frequency response of the interaction, across the frequency range, of the impedance of the output device and output impedance, or impedance of the output device and audio input device, are identified in various examples, This is achieved, at least in part, by measuring the impedance / frequency response of each item appearing in first drop-down menu 410 and each impedance or audio input device appearing in the second drop-down menu 420. These measurements on specific devices are carried out using known techniques.

[0102] Turning to the earpad category, in the example shown in Figure 4, this is shown to include a two-way toggle switch 430. This allows a user to toggle between an artificial leather earpad type and a soft cloth earpad type. In other examples, further options are also able to be provided. Using the technique set out above, the difference from a default across the frequency range, such as an artificial leather earpad type, is able to be identified. A negative to this difference, either independent from or in combination with a negative of the impedance category, is able to be identified. Again independent of or in combination with the negative of the impedance category, in some examples a relevant one or more filters that produce the negative or a negative reproduction of the relevant frequency response is able to be identified based on the user input. Either the relevant difference, frequency response thereof, negative, or relevant one or more filters are stored in a library, database or look-up table, the respective entry of which is read based on which earpad type is selected and / or the input(s) for the impedance category.

[0103] As the user makes selections in one or both of the two categories, the relevant equalisation filtering is arrived at based on those selections. This is then applied to the audio output signal passed from the audio input device 14 to the output device 18. This aims to restore a flat frequency response across the frequency range in the signal received by the speaker of the output device and / or audible to a user of the output device.

[0104] Figure 5 shows an alternative arrangement to that shown in Figure 4 for the impedance category. Figure 5 shows an example user interface 500. This has an example first drop-down menu 510 that provides the same functionality as the example first drop-down menu 410 of the example shown in Figure 4. In place of the second drop-down menu 420 shown in the example of Figure 4, the example user interface of Figure 5 has an example rotary dial 520. This has markings in the example of Figure 5 of 0, 10, 20, 40, 60, 80 and 100. These represent impedance values in Ohms. In a similar manner to choosing an item in a drop-down menu, a user is able to select the relevant impedance. This can either be on a continuous scale or discrete values, such as incrementing by 0.25 Ω, 0.5 Ω, 1 Ω, 5 Ω, 10 Ω or 20 Ω. Regardless of whether a continuous scale or discrete values, the input provided by this is treated in the same way by the audio processor as described above in relation to the second drop-down menu in the example of Figure 4.

[0105] Further alternatives also exist. The first drop-down menu 410, 510 is replaced by a toggle or other input types in some examples. The second drop-down menu 410 or rotary dial 520 is able to be replaced by a toggle, slider, numerical value input, range selection (such as 0 Ω to 20 Ω, 20 Ω to 40 Ω, 40 Ω to 60 Ω, 60 Ω to 80 Ω and 80 Ω to 100Ω) or other input types. The difference between any discrete values or ranges used is, in some examples, set so as not to create a difference in distortion of more than 1 dB from one setting to the next.

[0106] The choice between a continuous scale and discrete values is a choice between greatest accuracy compared to storage capacity used or processing requirements depending on whether a look-up table or the like is used or calculations are conducted in order to arrive at the final equalisation filter(s).

[0107] As a further alternative, the equalisation filters are able, in some examples, to be in the form of a curve or set of curves (such as a curve equation or a set of curve equations that are able to be applied).

[0108] Considering Figure 6, this shows an example process applied to implement an example audio processor. At step S600 shown in the example process of Figure 6, the audio input signal is received. In various examples, this is received by the audio processor 16, typically from the audio input device 14.

[0109] At step S610, one or more identified frequency responses are received. In some examples, this is received by the audio processor 16, such as by input at a user interface 400, 500 or by an automated process. While in the example process set out in Figure 6, this step is shown as occurring after step S600. In other examples, however, step S610 can happen before or contemporaneously with step S600.

[0110] Following receipt of the one or more identified frequency responses, at step S620, equalisation filtering is generated. This involves one or more of the calculation and read operations of a library, database or look-up table set out above.

[0111] The equalisation is then applied to the audio input signal at step S630. This provides an audio output signal that includes the audio input signal with the equalisation filtering applied. This is provided to the output device 18 in various examples.

[0112] In some examples, a non-transitory computer-readable medium, such as a hard-drive, solid-state drive or some other form of storage medium, is provided as generally illustrated at 1000 in Figure 7. This is capable of holding or storing a computer program 1001. The computer program includes instructions, typically in the form of computer code, that, when implemented on a computing device, such as by executing the program, cause an apparatus to perform one or more of the methods and processes set out above or to provide the audio processor set out above. The apparatus may be a computing device such as a chip, server or some other form of computer.

[0113] The above examples are to be understood as illustrative examples. Further examples are envisaged. It is to be understood that any feature described in relation to any one example may be used alone, or in combination with other features described, and may also be used in combination with one or more features of any other of the examples, or any combination of any other of the examples. Furthermore, equivalents and modifications not described above may also be employed within the scope of the accompanying claims.

Examples

Embodiment Construction

[0052]An arrangement implementing an example audio processor is generally illustrated at 1 in Figure 1. The arrangement shown in Figure 1 includes example audio generation devices in the form of a microphone 10 and guitar 12. In other examples, the audio generation devices may take another form, such as a different instrument or audio generation device. These include synthesisers, streamed audio reproduced via an audio player, and record, cassette or CD decks.

[0053]The audio generation devices are connected to an audio input device 14 by standard connection means, such as audio cables. In various examples, the audio input device is an audio interface, such as a Focusrite ®< Scarlett ®< 2i2 ®< . Other audio input devices include other audio production equipment, such as an amplifier, MIDI device and mixer.

[0054]In some examples, the audio input device 14 is also an audio generation device, or is able to access stored (or itself stores) audio input signals. In such examples, separat...

FIELD OF THE INVENTION

[0001] The present disclosure generally relates to audio processing, typically concerning adapting the audible output, such as from headphones, to take account of the effect of the audio production configuration on the audible output.BACKGROUND

[0002] Significant effort is applied when building audio production equipment to keep the form of the output as close a replication of the source as possible. This keeps the output "transparent" or undistorted rather than adding "colour" or distortion.

[0003] Keeping an output or signal transparent allows effects to be applied to a signal that is as close to "true" or undistorted as possible rather than needing to try to address detrimental consequences of the production equipment with effects. However, achieving this is complex and is typically a balancing act between resources, whether that be cost or power or some other limiting factor, and optimised output.

[0004] Additionally, a single entity rarely has control of an entire audio production chain from audio creation to audio output. This means equipment outside of an entity's control is also able to affect the signal, the effects of which may not be able to be addressed by the entity's equipment.

[0005] Audio output is provided through headphones or by loudspeakers. These are connected to an output of an item of production equipment, such as an amplifier, audio processor, synthesiser, Musical Instrument Digital Interface (MIDI) device, mixer, (record, cassette, CD, etc.) deck, computer, or some other audio production equipment.

[0006] The audible audio output from an output device (e.g. a headphone set or loudspeaker) is coloured by the configuration of the output device. For wired connections, the audible audio output is also coloured by the connection to the production equipment from which the output device receives an audio signal. For headphones, this is typically a connection to a headphone amplifier, which is often, but not always, part of another piece of production equipment.

[0007] For some headphone sets the configuration can be changed, such as by changing the earpad types. This affects the audible output from these headphone sets due to the change in materials. Additionally, the interaction of impedance of a headphone set and the impedance of the output to which the headphone set is connected can cause the frequency response of the headphone set to be coloured.

[0008] Headphone sets are available with different impedances. Picking a headphone set with a different impedance from the impedance of the output to which it is connected assists with reducing the colouring caused by this interaction. Indeed, some have proposed a solution of ensuring the headphone impedance is at least five times or even eight times the impedance of the output to which the headphone set is connected to provide a more transparent sound. However, this requires access to multiple sets of headphones or causes undesired output quality if a user only has low impedance headphones available to connect to a high impedance output. Additionally, this cannot take account of other aspects of a headphone set configuration altering the audible output, such as the type of earpad used, which is thus most clearly addressed by changing earpad type. However, this can affect user comfort or have other detrimental effects.

[0009] Accordingly, there is a need to provide a more transparent audible output from headphones regardless of overall configuration of an audio production chain.SUMMARY OF INVENTION

[0010] According to a first aspect, there is provided an audio processor suitable for modulating audible output of an (audio) output device, the audio processor being configured to: receive an audio input signal; receive an identified frequency response of an output device configuration across a frequency range; generate equalisation filtering based on the identified frequency response; and provide an audio output signal to the output device, the audio output signal including the audio input signal with the equalisation filtering applied.

[0011] Rather than reconfiguring the output device, or changing the output device, this allows the same output device to be used by adapting the signal to take account of the frequency response of the output device.

[0012] Typically, output devices, such as loudspeakers and headphones are selected based on the reported quality of the audio output relative to the price bracket a user is applying. This means it is usually unclear what other production equipment the output device is to be used with, or at least that the user will wish to be able to use the output device with a variety of production equipment. The instinct is therefore for adjustments to be made to the hardware of the output device to tailor its frequency response. This is because it allows such a tailored frequency response to be applicable to all production equipment the output device could be used with and not subject to variable quality electronics or calibration of electronics. However, we have found that providing an audio processor that is able to account for the frequency response of the output device through filtering provides a more reliable and versatile functionality, and thus offers an improved experience for the user.

[0013] The terminology "audible output" is intended to mean the output able to be heard by a user, such as the audio output received by a user. This is intended to be more than simply the audio output of a device. This is because audio that is output by a device, or at least, what is received by a user, is subject to interference or distortion by the environment between where the audio is output and where it is received. It is possible to measure audible output of a device as well as to identify intended audio output.

[0014] The output device is able to be any form of audio output device, including a loudspeaker. Typically, however, the output device is a headphone set, such as a pair of headphones or a single headphone. The terms "headphone set", "pair of headphones", "headphone pair", "single headphone", "headphones" or "headphone" are used interchangeably herein unless specifically stated otherwise.

[0015] The audio processor being configured to receive the identified frequency response may be an electronic frequency response or an audio frequency response. Further, the identified frequency response may be a transfer function. The audio processor may be further configured to identify the frequency response of the output device. This is in order for, or part of, the identified frequency response being received. Alternatively, the receiving of the identified frequency response may be due to the identified frequency response being provided to the audio processor, such as by a user, by identification in a library or from an external input, or by an algorithm configured to calculate a frequency response based on input parameters, such as output device frequency response or impedance and output impedance of the audio input device.

[0016] The frequency range may be a frequency range typically audible to (normal hearing) humans, such as 20 (or 15) Hertz (Hz) to 20,000 Hz, may be (an extended range such as) 10 Hz to 40,000 Hz or may be between 5 Hz and 100,000 Hz.

[0017] The identified frequency response of the output device may be any frequency response of the output device, or may include any attribute, property or influencing factor on the frequency response. Typically, the identified frequency response of the output device configuration includes an impedance of the output device configuration across the frequency range. Typically, the output impedance is stable at least over the normal audible frequency range of humans (i.e. 15 or 20 Hz to 20,000 Hz as set out above), and usually well beyond that range too. This is because the output stage is typically resistive.

[0018] While, when provided by a supplier, an impedance of an (audio) output device is typically quoted as a single value, such as 30 Ohms (Ω). Typically, that is the impedance at a particular frequency (usually a nominal number that represents the lowest point of a midband of an impedance curve of the output device). As such, while impedance is often reported as a single value in relation to audio equipment, a relevant impedance of any output device or any piece of audio equipment is typically a range of values, or varies, across a range of frequencies. This is because headphones and loudspeakers have a complex impedance consisting of resistive and reactive components.

[0019] This allows the equalisation to compensate for the impedance of the output device across the frequency range. The impedance of an audio output device (such as a headphone set or loudspeaker) is commonly not the same across the frequency range. Instead, the impedance response varies across the frequency range. This distorts an audio signal, and, due to the variation, is a variable distortion. This is often undesirable, especially when a user is looking for a transparent output. Accordingly, being able to compensate for the impedance variation significantly reduces the distortion of the output signal, providing an enhanced user experience and greater accuracy in the reproduction of the audio input signal. Further, this can reduce the amount of further processing a user may want to apply, thereby limiting processing power and equipment requirements.

[0020] The audio input signal may be received from an audio input device. This may result in the audio processor being further configured to receive an identified frequency response of the audio input device, and the equalisation filtering generation being further based on the identified frequency response of the audio input device, and thus based on the identified frequency response of the audio output device and the identified frequency response of the audio input device. Typically, but not always, this is when the identified frequency response of the output device includes an impedance of the output device configuration across the frequency range.

[0021] The combination of the equalisation being based on the frequency response, for example the impedance frequency response, of each of the audio input device and (audio) output device allows the equalisation to take account of the interaction between these frequency responses. While simply taking account of the frequency response of the audio output device will provide improved transparency of the audio output signal, there will be interaction between the two frequency responses. This means taking account of the interaction will further optimise the reduction in distortion.

[0022] It is noted that various forms of frequency response are referred to throughout. With the introduction above of two forms of frequency response, it is noted that various frequency responses may include a (mostly) resistive output impedance, such as of an audio input device, which does not significantly change with frequency, i.e. is a very "flat" frequency response; a frequency varying output device (e.g. headphone) impedance, which can be referred to as an "impedance curve"; and a resulting audible frequency response a user listens to or hears.

[0023] The source (e.g. the audio input device) output frequency response does not change significantly with frequency, hence being described as "flat". Additionally, output device designers attempt to provide output devices, such as headphones and loudspeakers, that have a user perceived flat frequency response. However, to achieve that, the frequency response is tailored to the specifics of the model of the device or the specific device itself, so is typically not flat. Typically, without intervention in accordance with aspects described herein, it is this last frequency response alters due to the impedance interaction between the audio input device and the output device.

[0024] The audio input device is intended to be an audio interface. The audio interface may, in some circumstances, also, or alternatively, be an audio generation device, such as a drum machine or musical instrument digital interface (MIDI) hardware, but this is not required. As such, the audio input device may be arranged in use to provide processing of an audio signal it receives from an audio generation device before providing that signal as an audio input signal to the audio processor of the first aspect. In some cases, the audio processor and audio input device may be part of, or modules of, the same piece of hardware.

[0025] Unlike audio output devices, audio input devices, when an audio interface or a piece of audio production equipment rather than purely an audio generation device that does not include a processor, the output impedance may be a constant value. By this it is intended to mean that the value is constant across the frequency range. There are, however, circumstances in which this output impedance may vary across the frequency range.

[0026] As a result of the output impedance of the audio input device and the impedance of the audio output device across the frequency range interacting, there is distortion of the audio signal output provided to the audio output device, or, in other words, the electrical signal passed to the audio output device. As such, if not taken into account, the identified frequency response of the audio input device, a default audio input device frequency response, such as a default output impedance, would likely be used to provide one part that causes the distortion.

[0027] Typically though, as set out above, the audio input device frequency response is provided. When the audio input device frequency response is provided as an output impedance of the audio input device across the frequency range, the audio processor may be further configured to calculate an audio signal distortion across the frequency range of an interaction of output impedance of the audio input device across the frequency range and the impedance of the output device configuration across the frequency range, the generated equalisation filtering being a reproduction of the calculated distortion as a negative.

[0028] This allows an imitation of the distortion to be generated to try to at least partially cancel out the distortion. The distortion will typically be an increase or decrease in signal strength or equivalent audio output volume from the true audio signal, and thus is an increase or decrease from zero. The equalisation filtering is intended to provide the opposite distorting effect on the audio signal to the distortion caused by the interaction of the output impedance and impedance of the output device. This replication thus diminishes the distortion.

[0029] The identified frequency response of the audio input device may be identified by the (specific type or model of) audio input device being provided as an input, such as by a user or as data from a source, including, but not limited to, user input or by data transmission from the audio input device or some other device connected to the audio processor and / or the audio input device. In these circumstances, the audio processor may include a lookup table or be able to access a lookup table, the lookup table including a correspondence between a plurality of audio input devices and frequency response. The audio process may then read a respective frequency response from the lookup table, receiving the respective frequency response as the identified frequency response. Typically, however, the identified frequency response of the audio input device may be received as an input provided to the audio processor. This reduces the storage capacity requirements and / or connectivity requirements of the audio processor, and limits data access and processing latency since it limits database access and read operations.

[0030] The input (of the identified frequency response) may be a value selected from a range or may be a section of the range selected from the range. The selection may be a user selection. This allows the user to select the appropriate frequency response for the audio input device being used, either by selecting a specific value (such as an impedance value), for example using a dial, slider or numerical input field, or by selecting a section of a range from a larger overall range (such as a sub-range of impedance values in an overall impedance range), for example using a drop-down menu, radio buttons, toggles, sliders or a (numerical) input field. The use of a specific value allows accurate selection of the frequency response and the use of a section of a range allows an approximate selection to be made. Since, if the extent of the section is suitably restricted, the differences in the frequency response at the maximum and minimum values in each section are not significant enough for it to affect the user experience. For example, each section may be a 20 Ω impedance range or a 10 Ω impedance range with the overall range extending from 0 Ω to 80 Ω or 100 Ω, such as 0 Ω to 20 Ω, 20 Ω to 40 Ω, 40 Ω to 60 Ω, 60 Ω to 80 Ω and 80 Ω to 100 Ω. The extent or resolution of each range (i.e. difference between the lower limit and upper limit of each range) is able to be based on what a user is able to perceive. For example, a resolution or range extent of 0.1 dB will provide audio differences that are imperceptible to a user. This means that having a set of ranges that cause an output to only vary by 0.1dB between adjacent ranges of impedance settings is less likely to be worthwhile. However, if a difference in output has a 0.5 dB to 1.0 dB resolution, that would be more worthwhile, since this resolution is on the threshold of audibility.

[0031] Additionally or alternatively to the identified frequency response of the output device configuration including an impedance of the output device configuration across the frequency range, the identified frequency response of the output device configuration may include an audible frequency response across the frequency range of the output device configuration. This allows the audio output signal to be modified to take account of factors that affect the audible frequencies emitted by the output device, such as materials used and placement of components, especially when these have been altered from a standard, default or designed configuration.

[0032] The audible frequency response across the frequency range may be a (pre)measured audible frequency response, but could be a calculated audible frequency response. The audible frequency response being measured allows a non-theoretical frequency response to be used. Since the audible frequency response will be affected by specific installation arrangements and manufacturing tolerances, using a theoretical audible frequency response is likely to be less accurate than a measured audible frequency response.

[0033] The output device may be any form of audio output device, such as a loudspeaker. Typically, however, the output device may be a headphone set and the output device configuration is the headphone set with one of a plurality of earpad types. In this case, the measured audible frequency response may be a measured audible frequency response of the headphone set with an identified earpad type of the plurality of earpad types. This allows the effect of earpad type on the audible audio output to be accounted for.

[0034] The earpad type may be identified by a user providing the earpad type, such as from a discrete list of earpad types or by providing data regarding earpad type. For a list of earpad types, the audio processor may have access to details of the audible frequency response for each earpad type, which is accessed to receive the identified frequency response.

[0035] The identified frequency response of the output device may be identified by the (specific type or model of) output device being provided as an input, such as by a user or as data from a source, including, but not limited to, user input or by data transmission from some other device connected to the audio processor or by identification or selection of a frequency response or impedance value. As such, typically, the identified frequency response of the output device configuration is received based on input provided to the audio processor.

[0036] Since the output device typically has a frequency response that varies across the frequency range, providing a single value for the frequency response may not be representative of the actual frequency response. This is because even if two output devices, such as devices from different manufacturers, have the same reported frequency response if reported as a single value, the response of those output devices across the frequency range may vary from each other. Thus, typically, the audio processor is further configured to read a stored database entry selected based on input provided to the audio processor, the stored database entry being the identified frequency response of the output device configuration.

[0037] The audio processor may have a database of, or have access to a database of, one or more frequency responses from which it may be able to read entries. This allows a simple lookup operation to be conducted instead of needing to calculate the frequency response. While this would be possible by measuring an amount of current drawn by the output device across a range of voltage frequencies, this is a more complex operation that can be desirable, but, at minimum, requires calibration. Accordingly, reading a stored database entry and using that as the identified frequency response is a less complex operation and provides a simplified operation. When a user provides input of the output device this allows the user choice of the selection avoiding an automated system mis-identifying the output device or frequency response.

[0038] The audio processor may be configured to generate the equalisation filtering (by one of): calculating a filter as a reproduction of a negative of the identified frequency response(s); or identifying a filter in a lookup table from the identified frequency response(s). Calculating the filter to apply allows dynamic creation of the filter, allowing lower storage capacity or access to lower storage capacity for the audio processor. On the other hand, identifying the filter using a lookup table allows lower processing power.

[0039] The filter may be one (or more) of: a biquad filter set; an infinite impulse response filter; a finite impulse response filter; or a parametric equalisation filter and (corresponding) centre value. These allow accurate reproduction of the frequency response(s) in order to counteract the distortion produced by the frequency response(s) by emulating the frequency response.

[0040] According to a second aspect, there is provided a method of modulating audible output of an output device, the method comprising: receiving an audio input signal; receiving an identified frequency response of an output device configuration across a frequency range; generating equalisation filtering based on the identified frequency response; and providing an audio output signal to the output device, the audio output signal including the audio input signal with the equalisation filtering applied.

[0041] The method according to the second aspect may apply any one or more of the functions set out above or may implement features of the audio processor according to the first aspect is able to perform or is configured to provide.

[0042] According to a third aspect, there is provided an audio processor suitable for modulating audible output of an output device, the audio processor being configured to: receive an audio input signal from an audio input device; receive an identified frequency response of an output device configuration across a frequency range and an identified frequency response of an audio input device across the frequency range; generate equalisation filtering based on the identified frequency response of the output device configuration and the identified frequency response of the audio input device; and provide an audio output signal to the output device, the audio output signal including the audio input signal with the equalisation filtering applied.

[0043] Typically, the identified frequency response of the output device configuration may include an impedance of the output device configuration across the frequency range and the identified frequency response of the audio input device may include an output impedance of the audio input device across the frequency range. In these circumstances, the audio processor may be further configured to calculate an audio signal distortion across the frequency range of an interaction of output impedance of the audio input device across the frequency range and the impedance of the output device configuration across the frequency range, the generated equalisation filtering being a reproduction of the calculated distortion as a negative.

[0044] According to a fourth aspect, there is provided a method of modulating audible output of an output device, the method comprising: receiving an audio input signal from an audio input device; receiving an identified frequency response of an output device configuration across a frequency range and an identified frequency response of an audio input device across the frequency range; generating equalisation filtering based on the identified frequency response of the output device configuration and the identified frequency response of the audio input device; and providing an audio output signal to the output device, the audio output signal including the audio input signal with the equalisation filtering applied.

[0045] According to a fifth aspect, there is provided an audio processor suitable for modulating audible output of an output device, the audio processor being configured to: receive an audio input signal; receive an identified audible frequency response of an output device configuration across a frequency range; generate equalisation filtering based on the identified frequency response; and provide an audio output signal to the output device, the audio output signal including the audio input signal with the equalisation filtering applied.

[0046] Typically, the output device may be a headphone set and the output device configuration may be the headphone set with one of a plurality of earpad types. The audible frequency response across the frequency range may be a measured audible frequency response. In such a case, the measured audible frequency response may be a measured audible frequency response of the headphone set with an identified earpad type of the plurality of earpad types.

[0047] According to a sixth aspect, there is provided a method of modulating audible output of an output device, the method comprising: receiving an audio input signal; receiving an identified audible frequency response of an output device configuration across a frequency range; generating equalisation filtering based on the identified audible frequency response; and providing an audio output signal to the output device, the audio output signal including the audio input signal with the equalisation filtering applied.

[0048] According to a seventh aspect, there is provided a computer program comprising instructions which, when executed, cause an apparatus to perform the method according to any one of the second, fourth and / or sixth aspects.

[0049] The computer program may be in the form of a plug-in, i.e. a software component that adds a specific functionality to another program. The other program may be a digital audio workstation (DAW). Alternatively, the computer program may be a standalone program that operates separately from other programs.

[0050] According to an eighth aspect, there is provided a non-transitory computer-readable medium comprising the computer program according to the seventh aspect.BRIEF DESCRIPTION OF DRAWINGS

[0051] Example audio processors and methods are described in detail below with reference to the accompanying drawings, in which: Figure 1 shows an example schematic arrangement including an example audio processor; Figure 2 shows an example output device impedance curve; Figure 3 shows an example distortion curve and equalisation filtering; Figure 4 shows a first example interface; Figure 5 shows a second example interface; Figure 6 shows an example process; and Figure 7 shows a schematic diagram of an example non-transitory computer-readable medium. DETAILED DESCRIPTION

[0052] An arrangement implementing an example audio processor is generally illustrated at 1 in Figure 1. The arrangement shown in Figure 1 includes example audio generation devices in the form of a microphone 10 and guitar 12. In other examples, the audio generation devices may take another form, such as a different instrument or audio generation device. These include synthesisers, streamed audio reproduced via an audio player, and record, cassette or CD decks.

[0053] The audio generation devices are connected to an audio input device 14 by standard connection means, such as audio cables. In various examples, the audio input device is an audio interface, such as a Focusrite ®< Scarlett ®< 2i2 ®< . Other audio input devices include other audio production equipment, such as an amplifier, MIDI device and mixer.

[0054] In some examples, the audio input device 14 is also an audio generation device, or is able to access stored (or itself stores) audio input signals. In such examples, separate audio generation devices need not be included.

[0055] The audio input device 14 is connected to an audio processor 16 and an audio output device 18. In the example shown in Figure 1, the audio output device is a headphone set. In other examples, the audio output device can be another form of audio output device, such as a loudspeaker.

[0056] The audio processor 16 is represented in the example shown in Figure 1 by a computer, but is able to be any form of audio processor. In various examples, however, the audio processor is a DAW, a plug-in in a DAW or a standalone (software) application. This allows the audio processor to be formed by a processor (of the computer) and to provide a user interface via the display of the computer. In other examples, however, the audio input device may also be the audio processor, or the audio processor may be provided by another device or computer program.

[0057] In the example shown in Figure 1, the audio input device 14 has an input connection to the audio processor provided by an input 142, such as a MIDI connection or USB connection, and an output connection from the audio processor provided by an output 162, such as a MIDI connection or USB connection. In some examples, the input and the output are provided by the same connection.

[0058] The audio processor 16 also has an audio output connected, typically by a cable, to the audio output device 18. An audio output signal is provided to the audio output device by this audio output.

[0059] Whether the audio output device 18 is a headphone set or loudspeaker, this will have an impedance. For headphones, the reported impedance is typically between about 8 Ω and 600 Ω. Headphones with an impedance between about 8 Ω and 50 Ω are considered "low impedance" headphones and headphones with an impedance above about 50 Ω are considered "high impedance" headphones.

[0060] Loudspeakers typically have a reported impedance between about 2 Ω and 16 Ω, with most speakers having an impedance of between about 4 Ω and 8 Ω.

[0061] Regardless of the reported impedance of a headphone set or loudspeaker, the impedance is not typically constant or level across the range of operable frequencies, including over the range of human audible frequencies. Instead, as shown in the example impedance curve 180 in Figure 2, the impedance varies across a frequency range.

[0062] In Figure 2, a plot of frequency (in Hertz, Hz) against impedance (in Ohms, Ω) is shown with a base 10 log scale for the frequency. This plot shows an impedance curve 180 for an example headphone set. Between 1 Hz and 10 Hz there is a slow, slightly upwardly curved increase in the impedance; between 10 Hz and about 80 Hz there is a curved, rapid, increase in the impedance, increasing about 12 times the amount the curve increased between 1 Hz and 10 Hz. From this peak, the impedance then drops down to a similar level as it is at 10 Hz by a frequency between about 1000 Hz and 1100 Hz. The impedance then increases in a shallow curve, increasing in gradient towards 100000 Hz. This plot is representative (if potentially simplified) of an impedance curve for a headphone set.

[0063] Loudspeakers also have well-known and / or recognisable typical impedance curves. The specifics of the impedance curve for a headphone set or loudspeaker is dependent on a number of factors and may vary significantly from the example plot shown in Figure 2.

[0064] It is not only the audio output device 18 that has an impedance, however. The audio output of the audio input device also has an impedance. This is commonly referred to as "output impedance". This corresponds to the amount of current able to be provided by the respective audio output. Devices considered to have a low output impedance usually, but not always, provide a high amount of current and devices considered to have a high output impedance usually, but not always, provide a low amount of current. What is, in fact, able to be provided is, however, typically, determined by a combination of output impedance, amplifier and / or power supply unit of / incorporated within the device. Regarding the effect of impedance, this is due to the impedance being able to be considered in some circumstances as a restriction on current flow.

[0065] When reported, output impedances of typical audio input devices can have a very large range from impedances of less than 1 Ω to impedances in excess of two factors of ten larger. While output impedance can vary across the frequency range, the output impedance is typically approximately constant, and level, across the frequency range shown in Figure 2 for the impedance curve.

[0066] Without an output device connected to its audio output, audio input devices are typically designed to have a flat frequency response. An example of a flat frequency response can be seen in the distortion plot of Figure 3 at 200 as a line lying along the horizontal axis origin (i.e. 0).

[0067] Figure 3 shows a plot of frequency against distortion (typically measured in decibels, dB). In addition to the example flat frequency response 200, this shows two other curves.

[0068] As noted above, the flat frequency response 200 is the frequency response of the audio input device when no output device is connected to the audio output. This is commonly referred to as "unloaded" due to the output device presenting a load to the audio input device when connected. Once connected, the frequency response changes. This changes due to the frequency response, including the impedance, of the output device across the frequency range and the interaction of that with the frequency response, including the output impedance, of the audio input device (and specifically, in various examples, the audio output of the audio input device to which the output device is connected).

[0069] Figure 3 shows an example distortion curve 210 of an audio signal across the frequency range when an output device is connected to an audio input device. The frequency response shape of the output device broadly corresponds to the example response shown in Figure 2, although may be shifted relative to the frequency range compared to the curve shown in Figure 2.

[0070] The example distortion curve 201 of Figure 3 shows a bell curve-like shape between 1 Hz and about 900 Hz with a peak between 10 Hz and 20 Hz. The curve then has a positive gradient as the frequency increases towards 100000 Hz.

[0071] When the distortion is at a level of around 1 dB or above, a listener is able to detect the distortion in the audible audio output. For trained sound engineers, the distortion can be identified at lower levels, but a 1 dB distortion level would generally be accepted as being audible. This is considered undesirable since it causes the volume of different sections of the frequency range to be different from each other when the intention is that the volume is constant across the frequency range (i.e. the frequency range audible to humans).

[0072] A further cause of distortion in a headphone set is distortion caused by the audible frequency response by the earpad material. There are several types of headphone, which include "in-ear", "over-ear" or "on-ear". For at least the latter two types, there are also "open back" and "closed back" (i.e. have a back or outward-facing portion that is, at least partially, open to the air or has a solid back) varieties.

[0073] Additionally, headphone sets can have earpads that can be changed. In on-ear and over-ear headphone sets, these comprise a cushion in a ring or (generally) circular shape. The size and shape of the cushion varies from model to model. The cushion provides a separation between a user's ear and the speaker in the headphone. This thereby forms a cavity that, in use, is defined by the cushion, headphone structure, speaker and the user's ear. The size, shape and composition of this cavity along with the corresponding characteristics of the headphone structure, speaker and user's ear all affect the audible sound across the frequency range of an audio output of the headphone set.

[0074] For in-ear headphone sets, there is still able to be a cavity defined by the portion of the headphone that sits within the user's ear, speaker or speaker grill and the user's ear. There may still be cushioning provided for such headphone sets with a form of earpad being provided by the portion of the headphone that sits within the user's ear, and which may be cushioning in some headphone sets.

[0075] The available variations in earpads allow different types of earpad to be used. For example, the earpad material may have an outer material or outer layer that is real or artificial leather, (soft) cloth or other fabrics, foam, plastic, silicone or other materials. As well as different outer materials or layers, earpads may be vented, such as by having one or more holes or perforations through which air is able to pass or shaped in a particular manner. The internal material or layers of an earpad may also vary, such as by foam pore size or compressibility, shape and / or composition. An earpad may also include one or more foils or other materials, or exclude one or more foils or other materials from the internal material(s). The earpad may further include a felt, foam or mesh layer between the cavity and the speaker within the headphone body.

[0076] There are several reasons to change the earpad type, such as comfort, number of expected users, and frequency and length of use. The earpad type (for example any of the combinations of material above) provides part of the output device configuration. This means that when the earpad type is changed, the output device configuration changes.

[0077] Due to the differences in composition and configuration of earpad types, this alters the effect of the earpad type on the cavity provided, and thus on the audible audio output from a headphone set. In view of this, headphone sets are designed for only a single earpad type or a theoretical average of some or all of the available earpad types for the headphone set. Should the user change the earpad type from the earpad type it is purchased with, or for which it is designed, this will thus alter the audible audio output of the headphone set. This alteration can be significant and can alter across the frequency range.

[0078] Sometimes this alteration can be desirable or sought out. For example, the Sennheiser ®< HD 490 Pro has two earpad sets, one for production of a velvet-like material, and one for mixing of a denim-like material. The production variety provides increased bass volumes, providing "warmth", whereas the mixing variety provides a flatter response up to about 1,000 Hz. While this alteration in the audible audio output is identified as beneficial in this situation, it is often undesirable.

[0079] It is possible to measure the difference in audible output produced by different earpad types. For an example headphone set, this measurement is achieved by using an ear coupler.

[0080] An ear coupler typically provides one flat plate per headphone. Each flat plate may have a rubber-like material and potentially an ear-like shape. This also includes a canal intended to replicate the human ear canal. One or more microphones are located within this canal.

[0081] In use a headphone is placed against the flat plate of the ear coupler in the orientation in which is mounted to the ear, sometimes held on with a jig to provide consistent placing when testing. Audio output is then played through the headphone that provides an output across the frequency range. This test is carried out with the headphone set for each earpad type (i.e. by having one earpad type attached to the headphone set for each test). By recording the output, it is then possible to identify the difference in audible audio output signal between different earpad types, and, in any case, the audible audio output across the frequency range for each earpad type.

[0082] In some examples, the difference in audible audio output signal of a headphone set with one earpad type compared to another or to a default earpad type may be similar to the distortion curve 210 shown in Figure 3. However, typically, this is different. Regardless, distortion still occurs.

[0083] As noted above, whatever the source of the distortion, this may be sought out or may be limited by changing output device configuration (whether that be due to a change in the device itself or just a change in the configuration). We have found that it is, instead, possible to address this distortion with "EQ" or equalisation, also referred to as equalisation filtering. This is due to a recognition of what contributes to the distortion.

[0084] We have recognised that the interaction causing the impedance related distortion is effectively due to a change in electrical dampening. What is meant by this is that, in effect, the output impedance of the audio input device and the impedance of the output device over the frequency range form a voltage divider. This divider provides different dampening behaviour depending on the frequency.

[0085] As noted above, the audio input device output impedance is (approximately) the same across the relevant frequency range, but can be different from one model of device to another; and the output device impedance varies with frequency with the characteristics of this variance changing from model to model. In the illustrative example of a voltage divider, it is possible to consider each of the audio input device and output device as two resistors (each representing the impedance of the respective device) connected in series. The resistor that represents the audio input device is a fixed resistance due to the consistent impedance across the relevant frequency range. In turn, the resistor that represents the output device is a variable resistance due to the impedance varying across the relevant frequency range. This means that when a voltage is applied across the opposing ends of these resistors (i.e. when the devices are in use) the voltage measurable between the mid-point between the resistors and one of the ends varies as the resistor representing the output device varies its resistance, but with the variation amount affected by both resistors.

[0086] This varying voltage is analogous to the distortion curve discussed above. This means that the characteristics of the distortion curve are affected by the output impedance of the audio input device as well as the impedance of the output device.

[0087] By providing EQ that is the negative of the distortion curve, an undistorted output is able to be provided. Since the audio output signal when passing from the audio input device to the output device is simply an electronic signal with a varying voltage, the provided EQ adjusts that voltage by application of a negative of the distortion curve, an undistorted output is able to be provided. In other words, this is because the negative curve, such as the example dashed line curve 220 shown in Figure 3, cancels the distortion that would otherwise be present.

[0088] To mimic or replicate a negative version of the distortion curve of the audible audio output signal, frequency response of the output device (such as due to its impedance) or the interaction of the frequency response of each of the output device and the audio input device (such as due to the impedance of the output device and the output impedance of the audio input device), EQ filtering is generated in use. We have identified this is possible to achieve when the contributions to the distortion are known, or at least the characteristics of the contributions. This is because, when the varying frequency response of the output device is known, it is possible to model the resulting distortion due to the contribution of a fixed frequency response. Once the varying frequency response and an example fixed frequency response are known, the distortion curve for that combination of parameters can be calculated. However, this means knowledge derived from multiple devices along the audio production chain is used. This has not previously been considered since, instead, focus has been on refining an optimising individual devices in the chain, and due to parties typically being limited to supplying products in one portion of the audio production chain only.

[0089] With this recognition, we have identified that it is possible for filtering to reproduce, as closely as possible, a negative version of the distortion curve. This can be achieved by use of at least one form of filter. These include the following forms of filter: a biquad filter set; an infinite impulse response (IIR) filter; a finite impulse response (FIR) filter; a parametric / semi-parametric / quasi-parametric equalisation filter with a centre value for the filter; low cut filter; high cut filter; low shelf filter; high shelf filter; bell curve filter; band pass filter; high pass filter; low pass filter; notch filter; bass tone filter; treble tone filter; and peak filter.

[0090] In use, the filter(s) or filter combination can be selected as appropriate based on the frequency response across the frequency range of the output device configuration in terms of, in a number of examples, the impedance and / or audible audio output. This can also be selected, in some examples, based on the frequency response across the frequency range of the audio input device, such as the impedance of the audio input device. Additionally, or alternatively, this can be selected, in some examples, due to its phase characteristics to equalise also any phase distortions. Typically, an FIR filter would be used for this. In various examples, this is able to be based, at least partially, on the interaction of the impedance across the frequency range of the output device configuration and the impedance across the frequency range of the audio input device.

[0091] The filter(s) or filter combination can be generated by an algorithm configured to create filters to meet the relevant criteria. Alternatively, an algorithm may select filters from a library, database or look-up table. In other examples, the appropriate filters to use for one or more output device configurations and, in some examples, one or more audio input device or audio input device frequency responses are stored in a library, database or look-up table. The relevant entry is able to be read from the library, database or look-up table on receipt of the relevant frequency responses, for example, by a library, database, or look-up table query.

[0092] To be able to apply appropriate filtering, identification of the characteristics for the various scenarios able to be modelled is needed. In terms of how this information is received by the audio processor, in some examples, the frequency response of the output device and, when relevant, the audio input device is identified automatically. In other examples, this information is identified or provided by a user or is based on information provided by the user. Examples of this are shown in Figure 4 and Figure 5.

[0093] Figure 4 shows a user interface 400 of the audio processor, in this case as (part of) an example plug-in for a DAW. This user interface has three user inputs in the example shown. In other examples, there can be more user inputs or less user inputs, and / or the user inputs are able to be provided in other forms.

[0094] The inputs shown in example of Figure 4 are in two categories. The categories are impedance and earpad type (shown in Figure 4 as "Earpad Material").

[0095] The impedance category has two inputs in the example shown in Figure 4. These are an example first drop-down menu 410 for headphone (identified in Figure 4 as "HP") model and an example second drop-down menu 420 for impedance. The earpad type category has one input in the example shown in Figure 4. This is a toggle switch 430 for earpad material.

[0096] In the example shown in Figure 4, the impedance category inputs allow a user to identify a headphone set model being used as an output device and an impedance of an audio input device. The first drop-down menu provides a list of options for the user to choose from in various examples. Similarly, in some examples, the second drop-down menu allows the user to either select a model (and optionally make) of an audio input device or an output impedance of the audio input device.

[0097] In various examples, the selections made in the first drop-down menu 410 and the second drop-down menu 420, identify a location in a library, database or look-up table. This location, in a number of examples, stores a predetermined frequency response of the interaction, across the frequency range, of the impedance of the headphone model and output impedance or audio input model (and optionally make). In some examples, the relevant one or more filters are then identified that produce a negative of the above frequency response, such as by generating the one or more filters or reading details of the one or more filters from a further library, database or look-up table.

[0098] In other examples, the location in the library database or look-up table provides details of the relevant one or more filters for each combination of selections from the first drop-down menu 410 and the second drop-down menu 420. This means that user selection from those menus provides the appropriate one or more filters to apply to produce a negative of a predetermined frequency response of the interaction, across the frequency range, of the impedance of the headphone model and output impedance or audio input device.

[0099] In some examples, the selection in the first drop-down menu identifies a location in a library, database or look-up table of a predetermined impedance of the selected headphone model across the frequency range and, together with the selection from the second drop-down, the interaction, across the frequency range, of the impedance of the headphone model and output impedance of the audio input device is calculated by the audio processor.

[0100] In various circumstances, one or more loudspeaker models may be included additionally or as an alternative to headphone models in the example first drop-down menu 410.

[0101] The predetermined impedance and / or frequency response of the interaction, across the frequency range, of the impedance of the output device and output impedance, or impedance of the output device and audio input device, are identified in various examples, This is achieved, at least in part, by measuring the impedance / frequency response of each item appearing in first drop-down menu 410 and each impedance or audio input device appearing in the second drop-down menu 420. These measurements on specific devices are carried out using known techniques.

[0102] Turning to the earpad category, in the example shown in Figure 4, this is shown to include a two-way toggle switch 430. This allows a user to toggle between an artificial leather earpad type and a soft cloth earpad type. In other examples, further options are also able to be provided. Using the technique set out above, the difference from a default across the frequency range, such as an artificial leather earpad type, is able to be identified. A negative to this difference, either independent from or in combination with a negative of the impedance category, is able to be identified. Again independent of or in combination with the negative of the impedance category, in some examples a relevant one or more filters that produce the negative or a negative reproduction of the relevant frequency response is able to be identified based on the user input. Either the relevant difference, frequency response thereof, negative, or relevant one or more filters are stored in a library, database or look-up table, the respective entry of which is read based on which earpad type is selected and / or the input(s) for the impedance category.

[0103] As the user makes selections in one or both of the two categories, the relevant equalisation filtering is arrived at based on those selections. This is then applied to the audio output signal passed from the audio input device 14 to the output device 18. This aims to restore a flat frequency response across the frequency range in the signal received by the speaker of the output device and / or audible to a user of the output device.

[0104] Figure 5 shows an alternative arrangement to that shown in Figure 4 for the impedance category. Figure 5 shows an example user interface 500. This has an example first drop-down menu 510 that provides the same functionality as the example first drop-down menu 410 of the example shown in Figure 4. In place of the second drop-down menu 420 shown in the example of Figure 4, the example user interface of Figure 5 has an example rotary dial 520. This has markings in the example of Figure 5 of 0, 10, 20, 40, 60, 80 and 100. These represent impedance values in Ohms. In a similar manner to choosing an item in a drop-down menu, a user is able to select the relevant impedance. This can either be on a continuous scale or discrete values, such as incrementing by 0.25 Ω, 0.5 Ω, 1 Ω, 5 Ω, 10 Ω or 20 Ω. Regardless of whether a continuous scale or discrete values, the input provided by this is treated in the same way by the audio processor as described above in relation to the second drop-down menu in the example of Figure 4.

[0105] Further alternatives also exist. The first drop-down menu 410, 510 is replaced by a toggle or other input types in some examples. The second drop-down menu 410 or rotary dial 520 is able to be replaced by a toggle, slider, numerical value input, range selection (such as 0 Ω to 20 Ω, 20 Ω to 40 Ω, 40 Ω to 60 Ω, 60 Ω to 80 Ω and 80 Ω to 100Ω) or other input types. The difference between any discrete values or ranges used is, in some examples, set so as not to create a difference in distortion of more than 1 dB from one setting to the next.

[0106] The choice between a continuous scale and discrete values is a choice between greatest accuracy compared to storage capacity used or processing requirements depending on whether a look-up table or the like is used or calculations are conducted in order to arrive at the final equalisation filter(s).

[0107] As a further alternative, the equalisation filters are able, in some examples, to be in the form of a curve or set of curves (such as a curve equation or a set of curve equations that are able to be applied).

[0108] Considering Figure 6, this shows an example process applied to implement an example audio processor. At step S600 shown in the example process of Figure 6, the audio input signal is received. In various examples, this is received by the audio processor 16, typically from the audio input device 14.

[0109] At step S610, one or more identified frequency responses are received. In some examples, this is received by the audio processor 16, such as by input at a user interface 400, 500 or by an automated process. While in the example process set out in Figure 6, this step is shown as occurring after step S600. In other examples, however, step S610 can happen before or contemporaneously with step S600.

[0110] Following receipt of the one or more identified frequency responses, at step S620, equalisation filtering is generated. This involves one or more of the calculation and read operations of a library, database or look-up table set out above.

[0111] The equalisation is then applied to the audio input signal at step S630. This provides an audio output signal that includes the audio input signal with the equalisation filtering applied. This is provided to the output device 18 in various examples.

[0112] In some examples, a non-transitory computer-readable medium, such as a hard-drive, solid-state drive or some other form of storage medium, is provided as generally illustrated at 1000 in Figure 7. This is capable of holding or storing a computer program 1001. The computer program includes instructions, typically in the form of computer code, that, when implemented on a computing device, such as by executing the program, cause an apparatus to perform one or more of the methods and processes set out above or to provide the audio processor set out above. The apparatus may be a computing device such as a chip, server or some other form of computer.

[0113] The above examples are to be understood as illustrative examples. Further examples are envisaged. It is to be understood that any feature described in relation to any one example may be used alone, or in combination with other features described, and may also be used in combination with one or more features of any other of the examples, or any combination of any other of the examples. Furthermore, equivalents and modifications not described above may also be employed within the scope of the accompanying claims.

Examples

Embodiment Construction

[0052]An arrangement implementing an example audio processor is generally illustrated at 1 in Figure 1. The arrangement shown in Figure 1 includes example audio generation devices in the form of a microphone 10 and guitar 12. In other examples, the audio generation devices may take another form, such as a different instrument or audio generation device. These include synthesisers, streamed audio reproduced via an audio player, and record, cassette or CD decks.

[0053]The audio generation devices are connected to an audio input device 14 by standard connection means, such as audio cables. In various examples, the audio input device is an audio interface, such as a Focusrite ®< Scarlett ®< 2i2 ®< . Other audio input devices include other audio production equipment, such as an amplifier, MIDI device and mixer.

[0054]In some examples, the audio input device 14 is also an audio generation device, or is able to access stored (or itself stores) audio input signals. In such examples, separat...

Claims

1. An audio processor suitable for modulating audible output of an output device, the audio processor being configured to: receive an audio input signal; receive an identified frequency response of an output device configuration across a frequency range; generate equalisation filtering based on the identified frequency response; and provide an audio output signal to the output device, the audio output signal including the audio input signal with the equalisation filtering applied.

2. The audio processor according to claim 1, wherein the identified frequency response of the output device configuration includes an impedance of the output device configuration across the frequency range.

3. The audio processor according to claim 1 or 2, wherein the audio input signal is received from an audio input device, the audio processor being further configured to receive an identified frequency response of the audio input device, and the equalisation filtering generation being further based on the identified frequency response of the audio input device.

4. The audio processor according to claim 3, wherein the identified frequency response of the audio input device includes an output impedance of the audio input device across the frequency range.

5. The audio processor according to claims 2 and 4, further configured to calculate an audio signal distortion across the frequency range of an interaction of output impedance of the audio input device across the frequency range and the impedance of the output device configuration across the frequency range, the generated equalisation filtering being a reproduction of the calculated distortion as a negative.

6. The audio processor according to any one of claims 3 to 5, wherein the identified frequency response of the audio input device is received as an input provided to the audio processor.

7. The audio processor according to claim 6, wherein the input is a value selected from a range or a section of the range selected from the range, the selection being a user selection.

8. The audio processor according to any one of the preceding claims, wherein the identified frequency response of the output device configuration includes an audible frequency response across the frequency range of the output device configuration.

9. The audio processor according to claim 8, wherein the audible frequency response across the frequency range is a measured audible frequency response.

10. The audio processor according to claim 8 or claim 9, wherein the output device is a headphone set and the output device configuration is the headphone set with one of a plurality of earpad types, the measured audible frequency response being a measured audible frequency response of the headphone set with an identified earpad type of the plurality of earpad types.

11. The audio processor according to any one of the preceding claims, wherein the identified frequency response of the output device configuration is received based on input provided to the audio processor.

12. The audio processor according to any one of the preceding claims, wherein the audio processor is configured to generate the equalisation filtering by one of: calculating a filter as a reproduction of a negative of the identified frequency response; or identifying a filter in a lookup table from the identified frequency response.

13. A method of modulating audible output of an output device, the method comprising: receiving an audio input signal; receiving an identified frequency response of an output device configuration across a frequency range; generating equalisation filtering based on the identified frequency response; and providing an audio output signal to the output device, the audio output signal including the audio input signal with the equalisation filtering applied.

14. A computer program comprising instructions which, when executed, cause an apparatus to perform the method according to claim 13.

15. A non-transitory computer-readable medium comprising the computer program according to claim 14.

Images