Methods and systems for generating harmonics, and amplitude-ratio harmonic unit for virtual bass systems.

Abstract

A method and corresponding system for generating harmonics based on an input signal are provided. The method includes (S1) obtaining one or more frequency band signals from the input signal; generating (S2) a first amplitude scaling signal comprising one or more harmonic components based on at least one or each of the frequency band signals, the amplitude of the first amplitude scaling signal being proportional to the amplitude of the input signal; generating (S3) a normalized signal based on at least one or each of the frequency band signals, the amplitude of the normalized signal being independent of the amplitude of the input signal; and for at least one or each of the frequency band signals, multiplying the normalized signal by the corresponding first amplitude scaling signal to generate a second amplitude scaling signal comprising one or more harmonic components; and for at least one or each of the frequency band signals, generating (S4) an output signal based on the first amplitude scaling signal comprising one or more harmonic components and the second amplitude scaling signal comprising one or more harmonic components.

Description

Technical Field

[0001] The proposed technology generally relates to audio processing, and more specifically to a method and system for generating harmonics, a system for generating virtual bass signals (also known as psychoacoustic bass perception signals), an amplitude proportional harmonic unit for a virtual bass system, an audio processing system and a corresponding overall audio system, a computer program and a computer program product, and a corresponding device. Background Technology

[0002] Due to physical limitations, small amplifiers / drivers cannot efficiently reproduce low-frequency audio content; this can lead to an undesirable loss of bass feel (which plays a major role in overall perceived audio quality) and, in the worst case, a complete loss of musical content.

[0003] However, due to a psychoacoustic illusion known as the "fundamental frequency vanishing phenomenon," the listener may experience the illusion of a fundamental frequency even if it is not physically reproduced—a phenomenon known as virtual bass, virtual bass enhancement, or psychoacoustic bass enhancement. This phenomenon is based on the observation that the perceived pitch of a set of harmonics (i.e., integer multiples of the fundamental frequency) is determined not only by the fundamental frequency of the harmonics but also by the greatest common divisor of the existing frequencies. Generally, the purpose of a virtual bass system is therefore to generate at least two harmonics within a frequency range that can be reproduced by a playback system, thereby giving the listener the sensation of improved bass.

[0004] It is well known that nonlinear devices (NLDs) can be used to generate harmonics. For example, a full-wave rectifier can be used to generate even harmonics, a half-wave rectifier can generate the fundamental frequency plus even harmonics, and a full-wave integrator can be used to generate both even and odd harmonics. A large number of different NLDs have been studied in the context of virtual bass systems (see, for example, reference [1]). However, a limitation of most NLDs is that they result in different low-frequency perception at low input signal levels compared to high input signal levels. This is because in most nonlinear devices, the amplitude of the generated harmonics is nonlinearly related to the input signal level, i.e., these harmonics are not proportional to the amplitude. To address this problem, reference [2] suggests introducing a level detector for normalizing the input into the harmonic generator, thereby allowing the free generation of any combination of odd and even harmonics and their amplitude relationship with the fundamental harmonic.

[0005] Figure 1This is a schematic diagram of a conventional (narrowband) virtual bass generation circuit including an NLD-based harmonic generator circuit. First, the input signal is processed through a low-pass filter (to prevent intermodulation distortion), the cutoff frequency of which is set to the upper limit of the desired low-frequency extension. Then, the low-pass signal is processed through an NLD to generate harmonic components. The NLD output is passed through a band-pass filter to shape the harmonics and remove intermodulation components below the physical range of the playback system. The harmonics are then added to the high-pass filtered input signal.

[0006] Figure 2 This is a schematic diagram illustrating a common (wideband) extension used in NLD-based virtual bass systems. For example, see [2]. Here, when there may be more than one strong low-frequency component at the input of the harmonic generator, the input spectrum is divided into small bands and a separate harmonic generator is assigned to each band in order to reduce intermodulation distortion generated during harmonic generation.

[0007] While existing technologies offer some possibilities for providing virtual bass or psychoacoustic bass enhancement signals, there remains a general need for new and improved developments regarding the generation of harmonics in audio systems. Summary of the Invention

[0008] The overall objective is to provide new and improved developments regarding the process of generating harmonics in audio systems.

[0009] The specific objective is to provide a system for generating harmonics.

[0010] Another objective is to provide a system for generating virtual bass signals, also known as psychoacoustic bass sensation signals.

[0011] The specific purpose is to provide an amplitude ratio harmonic unit for a virtual bass system.

[0012] Another objective is to provide an audio processing system.

[0013] Another objective is to provide a corresponding overall audio system.

[0014] Another objective is to provide a method for generating harmonics.

[0015] Another objective is to provide a computer program and a computer program product.

[0016] Another objective is to provide a corresponding device configured to perform a method for generating harmonics.

[0017] These and other objectives are achieved through embodiments of the proposed technology.

[0018] According to the first aspect, a system for generating harmonics is provided. The system essentially includes:

[0019] - Input, which is used to receive an input signal with a base frequency;

[0020] -At least two signal paths:

[0021] A first signal path is configured to receive the input signal and to generate a first amplitude proportional signal comprising one or more harmonic components based on the input signal, the amplitude of the first amplitude proportional signal being proportional to the amplitude of the input signal.

[0022] A second signal path, configured to receive the input signal, includes at least one signal processing block configured to: i) generate a normalized signal based on the input signal, the amplitude of which is independent of the amplitude of the input signal; and ii) multiply the normalized signal by the first amplitude scaling signal to generate a second amplitude scaling signal including one or more harmonic components; and

[0023] - A combiner configured to generate an output signal based on a first amplitude scaling signal comprising one or more harmonic components and a second amplitude scaling signal comprising one or more harmonic components.

[0024] According to a second aspect, a system for generating a virtual bass signal, also known as a psychoacoustic bass perception signal, is provided, wherein the system for generating the virtual bass signal includes the system for generating harmonics as described in the first aspect.

[0025] According to a third aspect, an arrangement is provided that includes multiple instances of the system for generating harmonics according to the first aspect, wherein these multiple instances of the system for generating harmonics can be arranged in parallel and / or in series.

[0026] According to a fourth aspect, an audio processing system is provided, comprising a system for generating harmonics according to a first aspect, a system for generating virtual bass signals according to a second aspect, and / or an arrangement according to a third aspect.

[0027] According to a fifth aspect, an audio system is provided, which includes the audio processing system described in the fourth aspect.

[0028] According to a sixth aspect, a method for generating harmonics based on an input signal having a fundamental input frequency is provided. The method essentially includes:

[0029] - Obtain one or more frequency band signals from the input signal;

[0030] - Generate a first amplitude scaling signal comprising one or more harmonic components based on at least one or each of these frequency band signals, wherein the amplitude of the first amplitude scaling signal is proportional to the amplitude of the input signal;

[0031] - Generate a normalized signal based on at least one or each of these frequency band signals, the amplitude of which is independent of the amplitude of the input signal, and multiply the normalized signal by a corresponding first amplitude scaling signal for at least one or each of these frequency band signals to generate a second amplitude scaling signal including one or more harmonic components.

[0032] - For at least one or each of these frequency band signals, an output signal is generated based on the first amplitude ratio signal including one or more harmonic components and the second amplitude ratio signal including one or more harmonic components.

[0033] According to a seventh aspect, a computer program is provided for generating harmonics based on an input signal having an input frequency when executed by a processor, wherein the computer program includes instructions that, when executed by the processor, cause the processor to perform the method according to a sixth aspect.

[0034] According to the eighth aspect, a computer program product is provided, the computer program product including a non-transitory computer-readable medium thereon storing the computer program according to the seventh aspect.

[0035] According to a ninth aspect, an apparatus configured to perform the method according to a sixth aspect is provided.

[0036] For example, the proposed technique can use a unique combination of appropriately configured components to generate a first amplitude proportional harmonic series and a second amplitude proportional harmonic series, which are mixed together in a well-balanced manner and are typically proportional to the amplitude of the input signal.

[0037] Furthermore, for example, the proposed amplitude proportional harmonic generator technology can be configured to provide fractional harmonics in addition to natural (even / odd) harmonics.

[0038] A related aspect of the invention relates to a method in which natural harmonic generation and fractional harmonic generation are arranged in parallel branches and subsequently combined.

[0039] Typically, the proposed techniques offer new and improved ways to generate harmonics in audio systems.

[0040] Other advantages will be understood when reading the following detailed description of non-limiting embodiments of the present invention. Attached Figure Description

[0041] The embodiments and their further objects and advantages can be best understood by referring to the following description taken in conjunction with the accompanying drawings, in which:

[0042] Figure 1 This is a schematic diagram of a conventional (narrowband) virtual bass generation circuit that includes a harmonic generator circuit.

[0043] Figure 2 This is a schematic diagram illustrating a common (wideband) extension used in virtual bass systems.

[0044] Figure 3 It is a schematic block diagram illustrating a simplified example of an audio system.

[0045] Figure 4 This is a schematic diagram illustrating an example of a system for generating harmonics according to an embodiment.

[0046] Figure 5A This is a schematic diagram illustrating a non-limiting example of a harmonic generator (also known as an amplitude proportional harmonic unit (APHU)).

[0047] Figure 5B This is a schematic diagram illustrating another non-limiting example of APHU.

[0048] Figure 5C This is a schematic diagram illustrating yet another non-limiting example of APHU.

[0049] Figure 5D This is a schematic diagram illustrating an example of a submodule called a Symbol Sequence Unit (SSU), which helps to select and generate a new square waveform with a certain period based on the pure tone input and the configuration of the frequency divider and / or multiplier.

[0050] Figure 5E This is a schematic diagram illustrating an example of how an APHU can be configured to generate fractional harmonics using an SSU that includes a frequency divider and / or a frequency multiplier.

[0051] Figure 5F This is a schematic diagram illustrating a specific example of how APHU can be configured to generate fractional harmonics by using an SSU that includes a divider after the SIGN() function.

[0052] Figure 5G This is a schematic diagram illustrating an example of how APHU can be configured to generate fractional harmonics using NLD and SSU.

[0053] Figure 5HThis is a schematic diagram illustrating an example of how an APHU can be configured to generate fractional harmonics by using an additional NLD after a filtered sequence of symbols.

[0054] Figure 6 This is a schematic diagram illustrating an example of how a harmonic series can be generated using an APHU unit.

[0055] Figure 7 This is a schematic diagram showing that other types of harmonic series are also possible.

[0056] Figure 8A This is a schematic diagram illustrating an example of extending a basic configuration to multiple parallel branches, which allows for additional flexibility in the construction of harmonic series.

[0057] Figure 8B This is a schematic diagram illustrating an example of a cascaded arrangement of multiple system instances (e.g., implemented as APHU).

[0058] Figure 9 This is a schematic diagram illustrating an example of how a harmonic series can be shaped using the extension shown in Figure 8.

[0059] Figure 10 This is a schematic diagram illustrating an example embodiment of a system configuration in which the input signal is filtered by an input filter before the APHU processes the input signal.

[0060] Figure 11 This is a schematic diagram illustrating an example embodiment of a system configuration where the input signal is a stereo signal that is first converted into a mono signal.

[0061] Figure 12 This shows that the branch has been expanded using additional APHU processing. Figure 11 A schematic diagram of an example embodiment of the system configuration of the system.

[0062] Figure 13 This is a schematic diagram illustrating an example embodiment of a system configuration with N ≥ 2 parallel branches for generating harmonic series.

[0063] Figure 14 This is a schematic diagram illustrating how even and odd harmonics can be well mixed together using an APHU processing system.

[0064] Figure 15 This is a schematic diagram showing an example of a frame used for bass enhancement.

[0065] Figure 16 This is a schematic diagram illustrating an example of the two lowest harmonics above the cutoff frequency fc for the input frequency.

[0066] Figure 17 It is a demonstration Figure 16 A schematic diagram of the corresponding example shows the two lowest harmonics above the cutoff frequency and the harmonic at 3 / 2 of the fundamental frequency from the natural harmonic group.

[0067] Figure 18 It is a schematic diagram illustrating an example of how natural harmonic generation and fractional harmonic generation can be arranged in parallel branches and then combined.

[0068] Figure 19 This is a schematic diagram illustrating an example of a system with multiple parallel branches, where each branch maintains a different configuration of the APHU.

[0069] Figure 20 This is a schematic diagram illustrating an example of a computer implementation method according to an embodiment.

[0070] Figure 21 This is a schematic diagram illustrating an example of a method for generating harmonics based on an input signal having a fundamental input frequency. Detailed Implementation

[0071] In all the accompanying drawings, the same reference numerals are used for similar or corresponding elements.

[0072] refer to Figure 3 It may be helpful to start with an overview of the audio system; the diagram illustrates a simplified audio system. Audio system 100 essentially comprises an audio processing system 200 and a sound generation system 300. Generally, the audio processing system 200 is configured to process one or more audio input signals that may be associated with one or more audio channels. The filtered audio signals are forwarded to the sound generation system 300 to produce sound.

[0073] As mentioned above, in order to reproduce sound correctly in some audio systems, it is very important to generate harmonics from an input signal with a fundamental frequency.

[0074] According to the first aspect, a system for generating harmonics is provided. The system essentially includes:

[0075] - Input, which is used to receive an input signal with a base frequency;

[0076] -At least two signal paths:

[0077] A first signal path is configured to receive the input signal and to generate a first amplitude proportional signal comprising one or more harmonic components based on the input signal, the amplitude of the first amplitude proportional signal being proportional to the amplitude of the input signal.

[0078] A second signal path, configured to receive the input signal, includes at least one signal processing block configured to: i) generate a normalized signal based on the input signal, the amplitude of which is independent of the amplitude of the input signal; and ii) multiply the normalized signal by the first amplitude scaling signal to generate a second amplitude scaling signal including one or more harmonic components; and

[0079] - A combiner configured to generate an output signal based on a first amplitude scaling signal comprising one or more harmonic components and a second amplitude scaling signal comprising one or more harmonic components.

[0080] For example, a first signal path can be configured to generate a first amplitude-ratio signal having one or more natural harmonics, and a second signal path can be configured to generate a second amplitude-ratio signal having one or more natural harmonics or having one or more fractional harmonics.

[0081] In a specific example, the first signal path is configured to generate a first amplitude-ratio signal, such as one including an even-order harmonic series, and the second signal path is configured to generate a second amplitude-ratio signal, such as one including an odd-order harmonic series.

[0082] In another example, the first signal path is configured to generate a first amplitude scaling signal, such as including an even harmonic series or an odd harmonic series, and the second signal path is configured to generate a second amplitude scaling signal, such as including at least one fractional harmonic component.

[0083] For example, the second signal path may include:

[0084] The SIGN unit is used to generate a sequence of alternating symbols at a rate specified by the fundamental frequency of the input signal.

[0085] A filtering unit configured to filter the changing symbol sequence to generate the normalized signal, the amplitude of which is independent of the amplitude of the input signal, and

[0086] A hybrid frequency multiplier configured to multiply the normalized signal by the first amplitude ratio signal to generate the second amplitude ratio signal.

[0087] For example, the SIGN unit and the filtering unit can be configured to generate a normalized signal as one or more filtered symbol sequences, which are used as amplitude-independent estimates of the fundamental frequency of the input signal.

[0088] In a specific example, the second signal path further includes a frequency divider or multiplier arranged between the SIGN unit and the filter unit, the frequency divider or multiplier being used to generate a sequence of changing symbols with a certain period based on the pure tone input and the configuration / setting of the frequency divider or multiplier, wherein the filter unit is configured to receive the sequence of changing symbols to generate a normalized signal, thereby enabling the generation of fractional harmonics as part of a second amplitude-ratio signal when the normalized signal is multiplied by the first amplitude-ratio signal.

[0089] Optionally, the first signal path may include at least one signal processing block configured to generate a first amplitude ratio signal.

[0090] For example, the at least one signal processing block of the first signal path may include a nonlinear device NLD configured to participate in the generation of a first amplitude ratio signal.

[0091] As an example, an NLD may include an absolute value unit ABS(), which is used to generate a series of natural harmonics scaled proportionally to the input signal.

[0092] Alternatively, the first signal path is configured to allow only the input signal to pass through as a first amplitude scaling signal.

[0093] In a particular example, the system further includes a first gain multiplier configured to scale a first amplitude ratio signal and / or a second gain multiplier configured to scale a second amplitude ratio signal, wherein the combiner is configured to receive the scaled first amplitude ratio signal and / or the scaled second amplitude ratio signal as input to generate an output signal.

[0094] For example, the system can be implemented as a so-called amplitude proportional harmonic unit (APHU).

[0095] According to a second aspect, a system for generating a virtual bass signal, also known as a psychoacoustic bass perception signal, is provided, wherein the system for generating the virtual bass signal includes the system for generating harmonics as described in the first aspect.

[0096] More specifically, the proposed technology can therefore provide an amplitude proportional harmonic unit (APHU) for a virtual bass system.

[0097] According to a third aspect, an arrangement is provided that includes multiple instances of the system for generating harmonics according to the first aspect, wherein these multiple instances of the system for generating harmonics can be arranged in parallel and / or in series.

[0098] According to a fourth aspect, an audio processing system is provided, comprising a system for generating harmonics according to a first aspect, a system for generating virtual bass signals according to a second aspect, and / or an arrangement according to a third aspect.

[0099] According to a fifth aspect, an audio system is provided, which includes the audio processing system described in the fourth aspect.

[0100] For example, the proposed technique can use a unique combination of appropriately configured components to generate a first amplitude ratio (e.g., even order) harmonic series and a second amplitude ratio (e.g., odd order) harmonic series, which are mixed together in a well-balanced manner and are generally proportional to the amplitude of the input signal.

[0101] For example, the proposed technique can be used to generate virtual bass signals, also known as psychoacoustic bass enhancement or sensory signals. In particular, a system 100; 200 for generating virtual bass signals, also known as psychoacoustic bass sensory signals, is provided, wherein the system 100; 200 for generating virtual bass signals includes a system for generating harmonics.

[0102] Figure 4 This is a schematic diagram illustrating an example of an audio processing system including a system for generating harmonics according to an embodiment.

[0103] In this particular example, an audio processing system 200 is provided, which includes a system 210 for generating harmonics. The system 210 for generating harmonics is sometimes referred to as a harmonic generator and / or includes a harmonic generator. Furthermore, an overall audio system 100 is provided, which includes such an audio processing system 200.

[0104] For better understanding, the invention will now be described with reference to additional non-limiting examples.

[0105] Figure 5AThis is a schematic diagram illustrating a non-limiting example of an APHU unit. The APHU unit consists of two parallel tracks: the first includes an ABS() unit and a first gain multiplier, and the second includes a SIGN() unit, a filter module, a hybrid multiplier, and a second gain multiplier. The ABS() function generates even harmonics and ensures that the harmonic series is scaled proportionally to the input amplitude. The sequence of changing signs (+ / -1) generated as the output of the SIGN() function alternates at a rate specified by the input fundamental frequency. A level-independent estimate of the fundamental frequency is obtained by low-pass filtering this sequence. The normalized fundamental frequency is then multiplied by the harmonic series generated by the ABS() function, which results in a series containing odd harmonics. The even and odd harmonic series are then scaled separately (scaled "gain a" and "gain b") and summed together to form a balanced mix of odd and even harmonics.

[0106] In other words, the proposed technique uses a unique combination of appropriately configured components to generate an even-harmonic series and an odd-harmonic series (amplitude ratio), which are mixed together in a well-balanced manner and are generally proportional to the amplitude of the input signal.

[0107] Figure 5B This is a schematic diagram illustrating another non-limiting example of APHU. Figure 5B Examples are similar to Figure 5A For example, the ABS unit is replaced by another (or more general) amplitude-proportional nonlinear device (NLD), which is used with the filtered output of the SIGN() function.

[0108] Figure 5C This is a schematic diagram illustrating yet another non-limiting example of an APHU. In this example, the scaling factors "gain a" and "gain b" have been set to 1, effectively reducing the component count of the entire circuit because the gain multiplier can be removed. This effectively means that the even-order harmonic series and the odd-order harmonic series are added together without any prior scaling.

[0109] It may be useful to continue with a brief, non-restrictive discussion of specific insights and key considerations, which ultimately led to a new evolutionary architecture for harmonic generation.

[0110] Nonlinear devices (NLDs) are typically static, memoryless, nonlinear elements / units that are well-suited to handling transients. As a counterexample, a full-wave integrator, often used as a harmonic generator, inherently causes the output to lag relative to the input. This suboptimal transient behavior can be effectively heard as a "pump bass" in a particular track. Because the integration of the output peaks will reach the end of the cycle, there is a lag after the input.

[0111] A drawback of most NLDs is that they provide amplitude-independent harmonic shaping. Harmonics are typically not scaled by the level detector as in reference [2].

[0112] The inventors have recognized that it may be beneficial to use filtered symbol sequences to obtain input amplitude-independent estimates of the fundamental frequency, which, when combined with NLDs, can be used to generate new harmonics relative to the harmonics generated by the NLDs (full-wave rectifiers generate even harmonics with a fixed amplitude relationship to the fundamental frequency).

[0113] Typically, full-wave rectifiers are simple to implement, but only generate even harmonics. This corresponds to a perceived pitch that is a full octave higher than the target perceived pitch.

[0114] In reference [3], the problem is solved by adding a soft limiting circuit to generate odd harmonics.

[0115] The inventors have realized that it might be better to obtain odd harmonics by multiplying the amplitude-proportional even harmonics by the amplitude-independent (normalized) estimate of the fundamental wave. As a result, a series of amplitude-proportional odd harmonics is generated.

[0116] It may be desirable to generate a harmonic family that will not depend on the amplitude of the input signal (beyond typical scaling). This can be a problem for static nonlinear systems. To address this issue, a so-called level detector is typically used, which can scale the output of the NLD.

[0117] In clear contrast, the inventors here recommend a method for expanding harmonic components into harmonics naturally generated by the NLD in a manner that keeps the amplitude of the resulting harmonic series constant with respect to the input signal. The concept of amplitude-proportional harmonics corresponds to all harmonics scaling proportionally with the input amplitude.

[0118] In the prior art, amplitude is preserved by means of envelope tracking in each part.

[0119] In clear contrast, the Amplitude Proportional Harmonic Unit (APHU) naturally generates a full (unrestricted) amplitude proportional harmonic series, which can optionally be shaped / limited by filtering later.

[0120] The principle behind fundamental wave vanishing is based on the importance of using odd and even harmonics to produce a clean and deep bass feel. For example, assuming a pure tone input, one might expect to generate harmonics where the odd and even harmonics have roughly the same harmonic gain distribution (timbre), the harmonic amplitude is proportional to the input amplitude, and there is very little intrinsic signal "memory" to provide good transient response and control over timbre.

[0121] References [1], [4] and [5] provide a comprehensive review and analysis of all currently available methods for low-frequency sound reproduction, and it should be noted that these references do not mention or relate any scheme similar to the inventive solution proposed in this patent application.

[0122] Figure 5D This is a schematic diagram illustrating an example of a submodule called a Symbol Sequence Unit (SSU), which helps to select the generation of a new square waveform with a certain period based on the pure tone input and the configuration of the frequency divider and / or frequency multiplier. The APHU can be configured in this way to generate different sets of amplitude-ratio fractional harmonic series. For more information on how to construct frequency dividers and frequency multipliers like these, see reference [6]. For example, frequency dividers and frequency multipliers can be implemented using standard components such as flip-flops, which are commonly used in logic design and circuits.

[0123] Figure 5E This is a schematic diagram illustrating how an APHU can be configured to generate fractional harmonics using an SSU that includes a frequency divider and / or a frequency multiplier. The net effect is that the APHU unit now generates an amplitude-ratio fractional harmonic series.

[0124] Figure 5F This is a schematic diagram illustrating a specific example of how the APHU can be configured to generate fractional harmonics by using an SSU that includes a divider after the SIGN() function. In this example, it shows how the APHU can be configured to generate fractional harmonics (e.g., 1 / 2f0, 3 / 2f0) by using a divider after the SIGN() function. Again, the net effect here is that the APHU unit now generates a series of amplitude-ratio fractional harmonics. It should also be noted that when generating fractional harmonics, possible operating modes include simply omitting the ABS() unit.

[0125] Optionally, such as Figure 5F As indicated by the upper signal path on the dashed line, the output signal can be combined with another amplitude scaling signal from an optional signal path configured to process the input signal in an optional further combiner.

[0126] Figure 5G This is a schematic diagram illustrating how an APHU can be configured to generate fractional harmonics by using NLD and SSU in combination with filtering steps. In the example where the NLD is an ABS unit, it will be combined with... Figure 5E Correspondingly, and Figure 5F This indicates that the NLD has been configured as a simple pass-through unity function.

[0127] Figure 5HIt demonstrates how a second NLD extension can be used after a filtered sequence of symbols. Figure 5G A schematic diagram of an example of APHU. NLD 2 does not need to have the same properties as NLD 1. An example could be NLD 2 configured as a polynomial-based NLD.

[0128] It should be understood that, Figures 5A to 5H Each of the different non-restrictive implementations of the harmonic generator APHU expressed in the text can be used alone or in combination with other signal processing blocks such as filters, compressors, and limiters.

[0129] To better understand the extensions used to generate fractional harmonics, the following additional information can be provided.

[0130] For example, for a pure tone input, the SIGN function in the APHU generates a square waveform (+1 / -1) with a period corresponding to the fundamental frequency of the pure tone. In a variant of the APHU unit, a standard frequency divider (or frequency multiplier) unit is applied after the SIGN function. This generates a new square waveform with a period based on the pure tone input and the frequency divider selection. The net effect is that the APHU unit now generates an amplitude-proportional fractional harmonic series.

[0131] Figure 5E , Figure 5F , Figure 7 and Figure 9 An example of an APHU variant used to generate fractional harmonic series is shown in the figure.

[0132] It should also be noted that the combination of fractional harmonics with even / odd natural (integer) harmonics can be beneficial in specific audio systems used to produce the desired bass feel.

[0133] exist Figure 5E The example shown is a typical integer / fractional amplitude scaling harmonic generation architecture.

[0134] Figure 5H An example of a general amplitude-scaled harmonic generation architecture using an NLD after a filtered sequence of symbols is shown. NLD 1 should be configured to generate scaled harmonics (e.g., by making it a single function or a full / half-wave rectifier). In the following branches, SSU and the filter H_APHU ensure that the input amplitude is not scaled to the output. NLD 2 can be configured more freely, for example, as a polynomial-based NLD. For more information on polynomial-based NLDs like this, see references [1] and [7].

[0135] From a methodological perspective, the proposed technique provides one or more novel procedures for generating harmonics.

[0136] According to the sixth aspect, a method for generating harmonics based on an input signal having a fundamental input frequency is provided.

[0137] refer to Figure 21 This method basically includes:

[0138] S1: Obtain one or more frequency band signals from the input signal;

[0139] S2: Generate a first amplitude proportional signal comprising one or more harmonic components based on at least one or each of these frequency band signals, wherein the amplitude of the first amplitude proportional signal is proportional to the amplitude of the input signal;

[0140] S3: Generate a normalized signal based on at least one or each of these frequency band signals, the amplitude of which is independent of the amplitude of the input signal; and for at least one or each of these frequency band signals, multiply the normalized signal by a corresponding first amplitude scaling signal to generate a second amplitude scaling signal including one or more harmonic components; and

[0141] S4: For at least one or each of these frequency band signals, an output signal is generated based on the first amplitude ratio signal including one or more harmonic components and the second amplitude ratio signal including one or more harmonic components.

[0142] For example, the first amplitude ratio signal may include one or more natural harmonics, and the second amplitude ratio signal may include one or more natural harmonics or one or more fractional harmonics.

[0143] In a specific example, the first amplitude scaling signal includes an even harmonic series, and the second amplitude scaling signal includes an odd harmonic series.

[0144] In another example, the first amplitude scaling signal includes an even harmonic series or an odd harmonic series, and the second amplitude scaling signal includes at least one fractional harmonic component.

[0145] As an example, the step of generating a first amplitude scaling signal based on at least one or each of the frequency band signals includes using the frequency band signal as an input to the nonlinear device NLD and using the output of the NLD as the amplitude scaling signal.

[0146] In a particular example, the step of generating a normalized signal based on at least one or each of these frequency band signals includes generating a symbol sequence based on said at least one or each of these frequency band signals and applying a filter to the symbol sequence to produce a filtered symbol sequence signal.

[0147] For example, the step of generating a normalized signal based on at least one or each of these frequency band signals further includes using a filtered symbol sequence signal as input to another NLD and using the output of another NLD as a normalized amplitude signal.

[0148] For example, the step of generating a normalized signal based on at least one or each of these frequency band signals further includes manipulating the rate of change of the symbol sequence.

[0149] Optionally, for at least one or each of these frequency band signals, the step of generating an output signal includes providing weighted versions of a first amplitude scaling signal and a second amplitude scaling signal and combining these weighted versions to generate an output signal.

[0150] In a specific example, a method for generating harmonics based on an audio input signal may include at least a subset of the following steps:

[0151] 1) Obtain one or more frequency band signals from the input.

[0152] 2) Generate a first amplitude scaling signal based on at least one or each of the frequency band signals, wherein the amplitude of the first amplitude scaling signal is proportional to the amplitude of the input, optionally through the following operation:

[0153] a. Using a frequency band signal as the input to the first (generalized) nonlinear device NLD, and

[0154] b. Use the output of the first (generalized) NLD as the amplitude scaling signal.

[0155] 3) Generate a normalized amplitude signal based on at least one or each of the frequency band signals, the amplitude of which is independent of the input amplitude, for example, through the following operation:

[0156] a. Generate a symbol sequence based on one of the frequency band signals.

[0157] b. Apply a filter to the symbol sequence to generate a filtered symbol sequence signal.

[0158] c. Optionally, the filtered symbol sequence signal can be used as input to the second (generalized) NLD, and

[0159] d. Use the output of the second (generalized) NLD (which may or may not have the same properties as the first generalized NLD) as the normalized amplitude signal.

[0160] 4) For at least one or each of the frequency band signals, multiply the first amplitude scaling signal by the normalized amplitude signal to generate a second amplitude scaling signal, the second amplitude scaling signal having a signal amplitude that responds proportionally to changes in the input amplitude, and

[0161] 5) For at least one or each of the frequency band signals, a harmonic signal is generated at least according to the second amplitude ratio signal.

[0162] In a particular example, the method may further include the step of generating an output signal based on the harmonic signals(s) and the input signal(s).

[0163] As an example, the step of generating an output signal can be performed by optionally combining a weighted version of (multiple) harmonic signals and an input signal (e.g., in the form of a first amplitude ratio signal).

[0164] Optionally, the generalized NLD may or may not generate harmonics based on its input frequency, and furthermore, at least one of the first or second (generalized) NLDs is an important (generalized) NLD, characterized in that the output of the NLD includes harmonics having the input frequency.

[0165] As mentioned, the first (generalized) NLD can have an output amplitude that is proportional to its input amplitude.

[0166] For example, at least one of the first or second (generalized) NLDs is an important (generalized) NLD, characterized in that: for a sinusoidal input signal to the NLD, the output of the NLD includes harmonics having the input frequency.

[0167] For example, a first (generalized) NLD can be defined by its properties as a nonlinear device for generating harmonics scaled proportionally to the fundamental amplitude.

[0168] In a specific example, the first (generalized) NLD is represented by a single mapping, a half-wave rectifier, or a full-wave rectifier.

[0169] In a sense, the proposed technique can be characterized by extracting at least one low-frequency band from the input signal. Based on this band, an amplitude scaling signal and a normalized amplitude signal are generated. The product of these two signals is obtained, and this product is used to generate a harmonic signal. The amplitude scaling signal can also be included in the harmonic signal. Based on the same band or several other bands, one or more other harmonic signals can be generated in a similar manner. All harmonic signals are then combined, and the result is further combined with the input signal to produce the output signal.

[0170] A further characteristic of an amplitude-proportional signal is that its amplitude responds proportionally to changes in the input amplitude.

[0171] In this paper, for the purpose of understanding the concept of "amplitude ratio," amplitude refers to the maximum offset of a waveform. Therefore, an amplitude ratio system can be defined by the property that a change in input amplitude results in a proportional change in output amplitude. Thus, the term "amplitude ratio" is commonly used in this paper; it should be remembered that it can also be called "amplitude magnitude ratio."

[0172] To further clarify, it can be stated that the amplitude normalization system has an output that is unaffected by changes in the input amplitude. In the special case where the amplitude of the input signal is 0, the amplitude normalization system is open to implementation-specific behavior.

[0173] For example, (multiple) normalized amplitude signals can be generated using a cascade of at least one symbol detection unit, an optional unit for manipulating the rate of change of the symbol sequence, and a filter as a mandatory final step. A characteristic of a normalized amplitude signal is that its amplitude does not respond to changes in the input amplitude.

[0174] In a specific example, prior to the multiplication step, at least one of the amplitude scaling signals or the normalized amplitude signals is further processed by an additional harmonic generation unit. Here, the harmonic generation unit is characterized in that, for a sinusoidal input, its output contains harmonics of the input frequency. When the amplitude scaling signal is amplified by such a harmonic generation unit, the harmonic generation should also possess amplitude scaling properties, such that its output responds proportionally to changes in its input signal.

[0175] Understandably, the symbol detection stage essentially removes any information about the input amplitude. Therefore, it is a natural consequence that the normalized amplitude signal is amplitude normalized. It is also understandable that multiplying the amplitude scaling signal by the normalized amplitude signal in multiple multiplication stages will result in yet another amplitude scaling signal.

[0176] Figure 6 This is a schematic diagram illustrating an example of how a harmonic series can be generated using APHU. In the top left diagram, the fundamental frequency f0 (dashed harmonic) is presented as an input to the ABS() nonlinearity, which in turn generates even harmonics (indicated by solid harmonic components). The bottom left diagram shows the result of multiplying the low-pass filtered sign function by the harmonic series generated by the ABS() nonlinearity. The net result of the multiplication is that each harmonic component is split into two odd harmonics. The right diagram shows the result of scaling the even harmonic series (generated by ABS()) and the odd harmonic series (generated by ABS()*Filt(Sign)) and adding them together.

[0177] Figure 7This diagram illustrates how other types of harmonic series can be made possible by simply altering the properties of the Symbol Sequence Unit (SSU) to make it correlated with other reasonable ratios of the fundamental frequency. In this example, the rate of symbol change is designed to match half of the fundamental frequency (i.e., f0 / 2). In this case, the result of multiplying ABS()*Filt(SSU) is the following harmonic series, where each of the even harmonics (from ABS) has been split into two components (still centered on the harmonic component), but this time only f0 is removed. The right figure shows the result of adding the two harmonic signals together. Since the distance between the harmonic components in the resulting harmonic series is f0 / 2, the pitch will be perceived as an octave lower (i.e., sub-bass).

[0178] Figure 8A This is a schematic diagram illustrating an example of an extension that allows for additional flexibility in the construction of harmonic series. This additional flexibility is achieved through parallel branches with APHUs, each performing a “spectral division” of the harmonic components generated by the NLD using different types of Symbol Sequence Units (SSUs). The final result is then obtained as a weighted sum of all harmonic series generated by each branch (using weights w1, w2, ..., wN).

[0179] It should be understood that the output obtained from one APHU may also be used as input to another APHU. In other words, as... Figure 8B As illustrated in the diagram, it is feasible to provide a series connection of two or more APHUs, where the output of one APHU is used as the input to another APHU. This can be considered as a form of APHU cascade.

[0180] Figure 9 This is a schematic diagram illustrating an example of how harmonic series can be shaped by combining different instances of APHU. This is using... Figure 8A Specific examples of the extensions are shown in the figure.

[0181] In this example, this uses Figure 5A and Figure 5F An example of a harmonic generator unit is shown in Figure A. In Figure A, the fundamental frequency f0 (dashed harmonic) is presented as a pair of harmonics. Figure 5A The APHU example shown in Figure 1 uses the ABS() nonlinear input, which in turn generates even harmonics (indicated by solid harmonic components). Figure B illustrates the result of “spectrum segmentation” of the harmonics (indicated by dashed lines), resulting from multiplying the low-pass filtered sign function by the harmonic series generated by the ABS() nonlinearity. (Source: [Original Source Name]) Figure 5A The output of the APHU example shown in Figure C is illustrated. In Figure D, the fundamental frequency f0 is presented as a response to... Figure 5FThe input to the APHU example is shown in Figure 1. In this case, there is no nonlinearity in the upper branch. Figure E shows the result of “spectral segmentation” of the input fundamental frequency (indicated by the dashed line) using SSU in this case, where the divider is set to divide by 2 to calculate f0 / 2. By linearly combining the results from two different instances of APHU, the harmonic series shown in Figure G is obtained, which is then a mixture of natural harmonics (integer multiples of the fundamental frequency) and fractional harmonics (fractional multiples of the fundamental frequency). It should be noted that Figure F corresponds to Figure E because of the zero multiplication from Figure D.

[0182] Figure 10 This is a schematic diagram illustrating an example embodiment of a system configuration in which an input signal is filtered by an input filter before being processed by the APHU unit. The harmonic series generated by the APHU unit is then filtered by an output filter, which may, for example, be used to further shape the harmonic components. The input and output filters may be IIR or FIR filters used for, for example, amplifying or attenuating certain frequencies. Examples of these filters may include high-pass filters, low-pass filters, band-pass filters, and / or shelving filters.

[0183] exist Figure 10 In certain example embodiments, the input filter or output filter can be constructed using a series of cascaded second-order filters. Figure 10 The system settings shown can be used for configurations involving mono, stereo, or multi-channel audio.

[0184] Figure 11 This is a schematic diagram illustrating an example embodiment of a system configuration where the input signal is a stereo signal that is first converted into a mono signal. The input (source) signal can be delayed as needed, but the delay "dsource" can also be set to 0. In the example embodiment, the source signal is filtered by the input source filter before it is combined (added) with a series of filtered and scaled harmonics already generated in the APHU branch. In a particular example embodiment, "gain virtual" can be set to 1.

[0185] Figure 12 This shows that the branch has been expanded using additional APHU processing. Figure 11The diagram illustrates an example embodiment of the system configuration. The input spectrum is divided into small sub-bands before being processed by two APHUs. This allows for a separate harmonic generator for each branch, reducing intermodulation distortion generated during harmonic generation, especially in cases where more than one strong low-frequency component may exist at the input of the harmonic generator. Therefore, two input filters, two APHUs, and two output filters process the mono signals from the two parallel branches, then combine these mono signals into a single signal (added together), which is then scaled by a scalar "gain virtualization". Next, the mono signal is converted to a stereo signal before being combined (added) with the input source signal or its filtered version.

[0186] Figure 13 This is a schematic diagram illustrating an example embodiment of a system configuration with N ≥ 2 parallel branches for generating the harmonic series. Because... Figure 12 The example diagram shows two branches, therefore Figure 13 An example diagram can be conceived as having more than two branches, all of which together contain the concept of having N≥2 parallel branches. Typically, each part is configured for a corresponding subband, and all branches are then tuned together to achieve a system with good overall performance.

[0187] To better understand the present invention, a more detailed but non-limiting discussion and disclosure of embodiments will now be given:

[0188] As previously mentioned, in a particular example embodiment, it is feasible to construct the core harmonic generator using the following key building blocks:

[0189] ABS / NLD function (full-wave rectifier): generates even harmonics and retains the level.

[0190] SIGN function: Generates odd harmonics with fixed levels, regardless of the input level (i.e., it normalizes the input).

[0191] The SIGN function with low-pass filtering: For a suitable cutoff frequency somewhere between the 1st and 3rd harmonics, the output will approximate the 1st harmonic (i.e., the fundamental frequency), but the level will be normalized.

[0192] Mixed multiplication: The product of two sine functions results in two sine waves with frequencies corresponding to the sum and difference of the input frequencies.

[0193] As previously mentioned, a Symbol Sequence Unit (SSU) can also be used to select and generate a new square waveform with a certain period based on the pure tone input and the configuration of the frequency divider and / or frequency multiplier.

[0194] like Figure 5AAs shown, the ABS() function, as an absolute value function, returns the amplitude-proportional even harmonics. A normalized fundamental frequency estimate is implemented using SIGN() and a low-pass filter. This estimate, multiplied by the ABS() output, generates an amplitude-proportional odd harmonic series. The generated even / odd harmonic series have comparable harmonic amplitude distributions and can be virtually smoothly blended using simple gain adjustments (gain_a / gain_b).

[0195] For example, to Figure 5A The APHU feed in the device generates a full harmonic series with a specific timbre by feeding a pure tone input. The timbre can be controlled by applying a harmonic shaping filter as a post-filtering operation (e.g., a bandpass filter).

[0196] In practice, it may be desirable to adjust one or more of the following blocks and / or settings (tuning) in real time, for example, while listening to tone sweeps and / or selected music material:

[0197] • The filter "H_APHU" (the filter in the APHU block after the SIGN() function),

[0198] • Scaling factors: "Gain a" and "Gain b"

[0199] • Post-processing harmonic shaping filter

[0200] This is to produce harmonics that leave a clean / good impression of pure sound for a given input frequency range. This can be, for example, part of a tuning process performed by an audio engineer.

[0201] In summary, the result is a unique combination of properly configured blocks and / or modules that possess one or more of the following qualities:

[0202] • Achieve amplitude proportionality in a simple and effective manner;

[0203] • Sound quality, achievable timbre (when combined with a harmonic output filter);

[0204] • Sound quality, good transient response (due to limited internal signal memory); and

[0205] • DSP code and runtime complexity, using, for example, cascaded dual second-order filters to implement different filtering steps (i.e., input filter, H_APHU, and output filter).

[0206] Refer again Figure 11 A single-band virtual bass architecture can be built around the APHU, for example, using the following non-limiting example setup with a simple filtering stage:

[0207] • Input filter:

[0208] • Shape the APHU input according to the desired bandwidth (e.g., 20Hz to 75Hz).

[0209] • Output filter:

[0210] • Shape the output harmonics according to the speaker characteristics and desired virtual bass tone (e.g., 75Hz to 140Hz).

[0211] • Input source filter:

[0212] • A high-pass filter used to remove low-frequency energy that speakers cannot reproduce.

[0213] In a particular example implementation, all filters (H_APHU, input filter, output filter, input source filter) can be implemented as cascaded bisecond-order parts. This is advantageous from the perspective of DSP implementation and runtime complexity, as well as limiting the number of tuning operations required for well-known high-level bisecond-order design parameters.

[0214] Figure 14 This is a schematic diagram illustrating how even and odd harmonics can be well mixed together using an APHU processing system.

[0215] from Figure 14 The diagram shows how to use even and odd harmonics. Figure 11 The APHU processing system is well integrated.

[0216] Pure tone loudspeaker response example:

[0217] ·20Hz input:

[0218] ·Even times: 40, 80, 120,...

[0219] ·Odd times: 20, 60, 100, 140,...

[0220] ·50Hz input:

[0221] ·Even times: 100, 200,...

[0222] ·Odd times: 50, 150, 250,...

[0223] Although the proposed innovation involves a general method for generating harmonics, in many cases it has been found to be best suited as part of a non-restrictive framework for bass enhancement.

[0224] Figure 15This is a schematic diagram illustrating an example of such a framework. The non-selectable block first includes blocks for selecting a frequency range of interest from which the fundamental component from which harmonics are expected to be generated exists. Typically, this is implemented using a low-pass filter or a band-pass filter. The output of the block (which contains a signal with a finite bandwidth) is passed to a harmonic generator, and the generated harmonics are recombined with another signal comprising all audible frequencies of the input signal. This creates an output signal comprising both the unaltered high-frequency component and the harmonics generated from the low-frequency component, allowing the listener to hear signals related to both the high-frequency fundamental and the low-frequency fundamental.

[0225] Some basic additions to the above framework are possible, but not strictly necessary for basic operation. More than one frequency band can be extracted in parallel with the first band and passed to another harmonic generator before being recombinated with higher frequencies. Additional processing (such as compressors, filters, etc.) can also be applied before or after the harmonic generator to further shape the signal. These can be placed both before / after the harmonic generator, or both before and after it. Similarly, additional processing can be applied to… Figure 15 The upper branch applies some additional processing. This could be a high-pass filter to remove frequency components with frequencies from which harmonics are generated. A compressor could also be placed here. It should be noted that many other options are possible without affecting the spirit of the harmonic generator itself. We will now leave this framework and continue describing the harmonic generator under the assumption that it is placed in a minimal version of the framework described.

[0226] When harmonics (which can be called natural harmonics) are generated at integer multiples of the fundamental frequency, the phenomenon of fundamental disappearance essentially indicates that the two lowest harmonics above the cutoff frequency fc for sound reproduction must have significant amplitudes in the output signal in order to express the illusion of a low-frequency fundamental.

[0227] The possible harmonics are then selected and spaced apart by the fundamental frequency, meaning that the higher the fundamental frequency, the greater the distance between the possible harmonics in the frequency range. For a fundamental frequency just below fc, this means that harmonics at almost 2 to 3 times fc must be generated. For lower fundamental frequencies, the spacing between harmonics is also smaller, so harmonics closer to the cutoff frequency can be used to provide the same effect.

[0228] Take a system with a 100Hz fc as an example. Then, for a fundamental frequency of 95Hz, harmonics at 190Hz and 285Hz are needed respectively. On the other hand, if the fundamental frequency is 45Hz, then harmonics at 135Hz and 180Hz are sufficient. Figure 16 This concept is illustrated by showing the two lowest harmonics above the cutoff frequency for the input frequency. It can be seen that the highest fundamental frequency below the cutoff frequency is associated with the highest harmonic frequency.

[0229] The inventors have discovered that this problem can also be solved by generating fractional harmonics, specifically harmonics at 3 / 2 of the fundamental frequency. Typically, this harmonic is combined with a harmonic at twice the fundamental frequency. Since these two notes represent the harmonics at 3 and 4 times the fundamental frequency, the fundamental frequency vanishing phenomenon allows the listener to perceive the fundamental frequency at half the input signal's fundamental frequency. However, in the musical scale, this is only an octave shift. The notes are perceived as the same but played an octave lower. Therefore, the musical content remains unchanged.

[0230] When performing this process, the harmonics at the third harmonic frequency can be omitted because two harmonic components are already available at lower frequencies. Therefore, the timbre of the harmonic series will be more focused on lower frequencies, resulting in a better listening experience. Furthermore, if a bandpass filter is placed after (multiple) harmonic generation units, the upper limit may be lowered.

[0231] Figure 17 yes Figure 16 The corresponding diagram shows the two lowest harmonics above the cutoff frequency and the harmonic at 3 / 2 of the fundamental frequency from all groups of natural harmonics. As can be seen, compared to... Figure 16 The highest harmonic component used here is in a lower frequency range. This comes at the cost of the perceived fundamental frequency being different from the input fundamental frequency.

[0232] For example, such as Figure 18 As illustrated in the diagram, this strategy can be implemented using two parallel APHU branches / units: one configured to generate integer harmonics of the fundamental frequency, and the other configured to generate fractional harmonics at at least 3 / 2 of the input frequency.

[0233] For the fundamental frequency in the higher range below the cutoff frequency, the primary interest lies with the fractional harmonics. In practice, these harmonics will also mix with the imperceptible fundamental frequency below fc or the low-level harmonics at twice the input, because the system's lower cutoff frequency and the various band-limited filters all have finite slopes, and thus provide a very small leakage signal outside their cutoff frequencies. This is usually sufficient to provide the illusion of a low-frequency fundamental frequency. Where appropriate, when the fundamental frequency is significantly below the cutoff frequency, it is only necessary to provide natural harmonics at full amplitude. Therefore, the upper band limit of the filter associated with the input of the fractional harmonic generator unit can be set higher than the upper band limit of the filter associated with the input of the natural harmonic generator unit.

[0234] However, in a general form, the same effect can be achieved by any method used to generate natural harmonics, fractional harmonics, or both. In this general form, the method is characterized by using two frequency bands derived from the input signal, wherein at least one frequency band is used as the input to a natural harmonic generator unit, and one frequency band is used as the input to a fractional harmonic generator unit. The fractional harmonic generator generates at least a component at 3 / 2 of the fundamental frequency. The generated natural harmonic series and / or fractional harmonic series can be further modified, for example, by a low-pass or band-pass filter. The natural harmonic signal and the fractional harmonic signal are combined into a harmonic signal. The harmonic signal can be further modified, for example, by a low-pass or band-pass filter. This harmonic signal is then combined with the input signal or another signal derived therefrom to produce an output signal. More specifically, the general form can also have an upper limit of the bandwidth of the filter associated with the input to the fractional harmonic generator unit that is higher than the upper limit of the bandwidth of the filter associated with the input to the natural harmonic generator unit.

[0235] Figure 19 This is a schematic diagram illustrating an example of a system with multiple parallel branches, where each branch maintains a different configuration of the APHU.

[0236] In a specific, non-restrictive example, the system includes three parallel branches that combine different APHU "types":

[0237] Branch 1: Generates only natural (even + odd) harmonics.

[0238] Branch 2: Generate a combination of fractional harmonics and natural harmonics.

[0239] Branch 3: Generate only fractional harmonics, such as only 11 / 2.

[0240] It will be understood that the methods and arrangements described in this article can be implemented, combined, and rearranged in a variety of ways.

[0241] For example, an apparatus is provided that is configured to perform the methods described herein.

[0242] For example, embodiments may be implemented in hardware or in software for execution by a suitable processing circuitry system, or a combination thereof.

[0243] The steps, functions, processes, modules, and / or blocks described herein can be implemented in hardware using any conventional techniques, such as discrete circuit or integrated circuit technology, including both general-purpose electronic circuit systems and special-purpose circuit systems.

[0244] Alternatively or as a supplement, at least some of the steps, functions, processes, modules and / or blocks described herein may be implemented in software such as computer programs so that they may be executed by a suitable processing circuitry system such as one or more processors or processing units.

[0245] Examples of processing circuitry systems include, but are not limited to, one or more microprocessors, one or more digital signal processors (DSPs), one or more central processing units (CPUs), video acceleration hardware, and / or any suitable programmable logic circuitry system, such as one or more field-programmable gate arrays (FPGAs) or one or more programmable logic controllers (PLCs).

[0246] It should also be understood that the general processing power of any conventional equipment or unit implementing the proposed technology can be reused. Existing software can also be reused, for example, by reprogramming existing software or by adding new software components.

[0247] Solutions based on a combination of hardware and software can also be provided. The actual hardware-software partitioning can be determined by the system designer based on many factors, including processing speed, implementation cost, and other requirements.

[0248] According to a seventh aspect, a computer program is provided for generating harmonics based on an input signal having an input frequency when executed by a processor, wherein the computer program includes instructions that, when executed by the processor, cause the processor to perform the methods described herein.

[0249] According to the eighth aspect, a computer program product is provided, the computer program product including a non-transitory computer-readable medium thereon storing the computer program according to the seventh aspect.

[0250] Figure 20 This is a schematic diagram illustrating an example of a computer implementation according to an embodiment. In this particular example, at least some of the steps, functions, processes, modules, and / or blocks described herein are implemented in computer programs 425, 435, which are loaded into memory 420 for execution by a processing circuitry system including one or more processors 410. The processors(multiple) 410 and memory 420 are interconnected to enable normal software execution. Optional input / output devices 440 may also be interconnected to the processors(multiple) 410 and / or memory 420 to implement the input and / or output of related data, such as input parameters(multiple) and / or derived output parameters(multiple).

[0251] The term “processor” should be interpreted in a general sense as any system or device capable of executing program code or computer program instructions to perform a particular processing, determination, or computational task.

[0252] The processing circuitry system, including one or more processors 410, is thus configured to perform well-defined processing tasks, such as those described herein, when executing computer program 425.

[0253] The processing circuitry system need not be dedicated to performing only the steps, functions, processes and / or blocks described above, but can also perform other tasks.

[0254] In a particular embodiment, computer program 425; 435 includes instructions that, when executed by processor 410, cause processor 410 to perform the tasks described herein.

[0255] The proposed technology also provides a carrier including a computer program, wherein the carrier is one of an electronic signal, an optical signal, an electromagnetic signal, a magnetic signal, an electrical signal, a radio signal, a microwave signal, or a computer-readable storage medium.

[0256] For example, software or computer programs 425; 435 may be implemented as computer program products, which are typically executed or stored on non-transitory computer-readable media 420; 430 (particularly non-volatile media). Computer-readable media may include one or more removable or non-removable memory devices, including but not limited to read-only memory (ROM), random access memory (RAM), optical disc (CD), digital versatile optical disc (DVD), Blu-ray disc, universal serial bus (USB) memory, hard disk drive (HDD) storage devices, flash memory, magnetic tape, or any other conventional memory device. A computer program can thus be loaded into the operating memory of a computer or equivalent processing device for execution by its processing circuitry.

[0257] When executed by one or more processors, the process flow proposed in this paper can be viewed as a computer flow. The corresponding apparatus can be defined as a set of functional modules, where each step executed by the processor corresponds to a functional module. In this case, the functional module is implemented as a computer program running on the processor.

[0258] Computer programs residing in memory can therefore be organized into appropriate functional modules that are configured to perform at least a portion of the steps and / or tasks described herein when executed by a processor.

[0259] Alternatively, these functional modules can be implemented primarily through hardware modules or alternatively through appropriate interconnections between related modules. Specific examples include one or more appropriately configured digital signal processors and other known electronic circuitry (e.g., discrete logic gates) and / or application-specific integrated circuits (ASICs) as previously mentioned, interconnected to perform specific functions. Other examples of available hardware include input / output (I / O) circuitry systems and / or circuitry systems for receiving and / or transmitting signals. The scope of software and hardware is purely a matter of implementation choice.

[0260] These embodiments are given by way of example only, and it should be understood that the proposed technology is not limited thereto. Those skilled in the art will understand that various modifications, combinations, and changes can be made to the embodiments without departing from the scope defined by the appended claims. Specifically, where technically feasible, different partial solutions from different embodiments can be combined in other configurations.

[0261] References

[0262] [1] NAY OO; WOON-SENG GAN: "Perceptually-Motivated Objective Grading of Nonlinear Processing in Virtual-Bass Systems", Journal of the Audio Engineering Society, November 2011.

[0263] [2]RONALDUS M.AARTS; STEPHANUS P.STRAETEMANS: "Circuit, Audio System and Method for Processing Signals, and a Harmonics Generator", U.S. Patent 6,111,960, August 29, 2000.

[0264] [3]HSIN-YUAN CHIU;TSUNG-FU LIN:“Virtual Bass Generating Circuit and Method”, US 2019 / 0238979A1, August 2019.

[0265] [4] ADAM J.HILL: “Analysis, Modeling and Wide-Area Spatiotemporal Control of Low-Frequency Sound Reproduction”, paper, University of Essex, January 2012, Chapter 6.

[0266] [5] Nay Oo: "Theoretical Analysis and Perceptual Evaluations on Nonlinear Devices in Virtual Bass System", PhD Dissertation, 2012, Nanyang Technological University.

[0267] [6] THOMAS C. BARTEE: “Digital Computer Fundamentals”, McGraw-Hill Kogakusha, edited by foreign students, 1972.

[0268] [7] Wee-Tong Lim and Nay Oo and Woon-Seng Gan: “Synthesis of Polynomial-Based Nonlinear Device and Harmonic Shifting Technique for Virtual Bass System”, IEEE International Workshop on Circuits and Systems, 2009.

Claims

1. A system for generating harmonics, the system comprising: Input, which is used to receive an input signal with a base frequency; At least two signal paths: A first signal path is configured to receive the input signal and to generate a first amplitude proportional signal comprising one or more harmonic components based on the input signal, the amplitude of the first amplitude proportional signal being proportional to the amplitude of the input signal. A second signal path is configured to receive the input signal. The second signal path includes at least one signal processing block configured to: i) generate a normalized signal based on the input signal, the amplitude of the normalized signal being independent of the amplitude of the input signal; and ii) multiply the normalized signal by the first amplitude scaling signal to generate a second amplitude scaling signal including one or more harmonic components. as well as A combiner configured to generate an output signal based on a first amplitude scaling signal comprising one or more harmonic components and a second amplitude scaling signal comprising one or more harmonic components. The second signal path includes: The SIGN unit is used to generate a sequence of alternating symbols at a rate specified by the fundamental frequency of the input signal. A filtering unit configured to filter the changed symbol sequence to generate the normalized signal, and A hybrid frequency multiplier configured to multiply the normalized signal by the first amplitude ratio signal to generate the second amplitude ratio signal.

2. The system for generating harmonics as described in claim 1, wherein, The first signal path is configured to generate a first amplitude ratio signal having one or more natural harmonics, and The second signal path is configured to generate the second amplitude ratio signal, which may have one or more natural harmonics or one or more fractional harmonics.

3. The system for generating harmonics as described in claim 2, wherein, The first signal path is configured to generate the first amplitude ratio signal, such as including an even harmonic series, and the second signal path is configured to generate the second amplitude ratio signal, such as including an odd harmonic series.

4. The system for generating harmonics as described in claim 2, wherein, The first signal path is configured to generate the first amplitude ratio signal, such as including an even harmonic series or an odd harmonic series, and the second signal path is configured to generate the second amplitude ratio signal, such as including at least one fractional harmonic component.

5. The system for generating harmonics as described in claim 1, wherein, The SIGN unit and the filter unit are configured to generate the normalized signal as one or more filtered symbol sequences, which are used as amplitude-independent estimates of the fundamental frequency of the input signal.

6. The system for generating harmonics as claimed in claim 1, wherein, The second signal path further includes a frequency divider or multiplier disposed between the SIGN unit and the filter unit, the frequency divider or multiplier being used to generate a sequence of changing symbols with a certain period based on the pure tone input and the configuration / setting of the frequency divider or multiplier, wherein the filter unit is configured to receive the sequence of changing symbols to generate the normalized signal, thereby enabling the generation of fractional harmonics as part of the second amplitude ratio signal when the normalized signal is multiplied by the first amplitude ratio signal.

7. The system for generating harmonics as claimed in claim 1, wherein, The first signal path includes at least one signal processing block configured to generate the first amplitude scaling signal, wherein the at least one signal processing block of the first signal path includes a nonlinear device (NLD) configured to participate in generating the first amplitude scaling signal, wherein the NLD includes an absolute value unit (ABS) for generating a natural harmonic series scaled proportionally to the input signal.

8. The system for generating harmonics as described in claim 1, wherein, The first signal path is configured to transmit the input signal as the first amplitude ratio signal.

9. The system for generating harmonics as claimed in claim 1, wherein, The system further includes a first gain multiplier configured to scale the first amplitude ratio signal and / or a second gain multiplier configured to scale the second amplitude ratio signal, wherein the combiner is configured to receive the scaled first amplitude ratio signal and / or the scaled second amplitude ratio signal as input to generate the output signal.

10. A system for generating a virtual bass signal, the system for generating a virtual bass signal comprising the system for generating harmonics as described in claim 1.

11. An arrangement for generating harmonics, comprising a plurality of instances of the system for generating harmonics as claimed in claim 1, wherein the plurality of instances of the system for generating harmonics are arranged in parallel, wherein, Each instance is for a single frequency band of the input signal.

12. A method for generating harmonics based on an input signal having a fundamental input frequency, the method comprising: One or more frequency band signals are obtained from the input signal; A first amplitude proportional signal comprising one or more harmonic components is generated based on at least one or each of the frequency band signals, wherein the amplitude of the first amplitude proportional signal is proportional to the amplitude of the input signal. A normalized signal is generated based on at least one or each of the frequency band signals, the amplitude of the normalized signal being independent of the amplitude of the input signal, and for at least one or each of the frequency band signals, the normalized signal is multiplied by a corresponding first amplitude scaling signal to generate a second amplitude scaling signal including one or more harmonic components. For at least one or each of the frequency band signals, an output signal is generated based on the first amplitude ratio signal including one or more harmonic components and the second amplitude ratio signal including one or more harmonic components. The step of generating a normalized signal based on at least one or each of the frequency band signals includes generating a symbol sequence based on the at least one or each of the frequency band signals and applying a filter to the symbol sequence to produce a filtered symbol sequence signal.

13. The method of claim 12, wherein, The first amplitude scaling signal includes one or more natural harmonics, and the second amplitude scaling signal includes one or more natural harmonics or one or more fractional harmonics.

14. The method of claim 13, wherein, The first amplitude ratio signal includes an even harmonic series, and the second amplitude ratio signal includes an odd harmonic series.

15. The method of claim 13, wherein, The first amplitude ratio signal includes an even harmonic series or an odd harmonic series, and the second amplitude ratio signal includes at least one fractional harmonic component.

16. The method of claim 12, wherein, The step of generating a first amplitude scaling signal based on at least one or each of the frequency band signals includes using the frequency band signal as an input to the nonlinear device NLD and using the output of the NLD as the first amplitude scaling signal.

17. The method of claim 12, wherein, The step of generating a normalized signal based on at least one or each of the frequency band signals further includes using the filtered symbol sequence signal as input to another NLD and using the output of the other NLD as the normalized signal.

18. The method of claim 12, wherein, The step of generating a normalized signal based on at least one or each of the frequency band signals further includes manipulating the rate of change of the symbol sequence.

19. The method of claim 12, wherein, For at least one or each of the frequency band signals, the step of generating an output signal includes providing weighted versions of the first amplitude scaling signal and the second amplitude scaling signal and combining these weighted versions to generate the output signal.

20. A non-transitory computer-readable storage medium having a computer program stored thereon, the computer program being used, when executed by a processor, to generate harmonics based on an input signal having an input frequency, wherein, The computer program includes instructions that, when executed by the processor, cause the processor to perform the method as described in claim 12.

Images