DETAILED ACTION
Claims 1-7 of the instant application are pending and have been examined.
Notice of Pre-AIA or AIA Status
The present application, filed on or after March 16, 2013, is being examined under the first inventor to file provisions of the AIA .
Information Disclosure Statement
The information disclosure statement (IDS) submitted on 12/16/2024 was filed. The submission is in compliance with the provisions of 37 CFR 1.97. Accordingly, the information disclosure statement is being considered by the examiner.
Claim Objections
Claim 3 objected to because of the following informalities: the limitation of “a supervised learning method” should read: the supervised learning method. Appropriate correction is required.
Claim 4 objected to because of the following informalities: the limitation of “a beam condition” should read: the beam condition. Appropriate correction is required.
Claim Interpretation
The following is a quotation of 35 U.S.C. 112(f):
(f) Element in Claim for a Combination. – An element in a claim for a combination may be expressed as a means or step for performing a specified function without the recital of structure, material, or acts in support thereof, and such claim shall be construed to cover the corresponding structure, material, or acts described in the specification and equivalents thereof.
The following is a quotation of pre-AIA 35 U.S.C. 112, sixth paragraph:
An element in a claim for a combination may be expressed as a means or step for performing a specified function without the recital of structure, material, or acts in support thereof, and such claim shall be construed to cover the corresponding structure, material, or acts described in the specification and equivalents thereof.
The claims in this application are given their broadest reasonable interpretation using the plain meaning of the claim language in light of the specification as it would be understood by one of ordinary skill in the art. The broadest reasonable interpretation of a claim element (also commonly referred to as a claim limitation) is limited by the description in the specification when 35 U.S.C. 112(f) or pre-AIA 35 U.S.C. 112, sixth paragraph, is invoked.
As explained in MPEP § 2181, subsection I, claim limitations that meet the following three-prong test will be interpreted under 35 U.S.C. 112(f) or pre-AIA 35 U.S.C. 112, sixth paragraph:
(A) the claim limitation uses the term “means” or “step” or a term used as a substitute for “means” that is a generic placeholder (also called a nonce term or a non-structural term having no specific structural meaning) for performing the claimed function;
(B) the term “means” or “step” or the generic placeholder is modified by functional language, typically, but not always linked by the transition word “for” (e.g., “means for”) or another linking word or phrase, such as “configured to” or “so that”; and
(C) the term “means” or “step” or the generic placeholder is not modified by sufficient structure, material, or acts for performing the claimed function.
Use of the word “means” (or “step”) in a claim with functional language creates a rebuttable presumption that the claim limitation is to be treated in accordance with 35 U.S.C. 112(f) or pre-AIA 35 U.S.C. 112, sixth paragraph. The presumption that the claim limitation is interpreted under 35 U.S.C. 112(f) or pre-AIA 35 U.S.C. 112, sixth paragraph, is rebutted when the claim limitation recites sufficient structure, material, or acts to entirely perform the recited function.
Absence of the word “means” (or “step”) in a claim creates a rebuttable presumption that the claim limitation is not to be treated in accordance with 35 U.S.C. 112(f) or pre-AIA 35 U.S.C. 112, sixth paragraph. The presumption that the claim limitation is not interpreted under 35 U.S.C. 112(f) or pre-AIA 35 U.S.C. 112, sixth paragraph, is rebutted when the claim limitation recites function without reciting sufficient structure, material or acts to entirely perform the recited function.
Claim limitations in this application that use the word “means” (or “step”) are being interpreted under 35 U.S.C. 112(f) or pre-AIA 35 U.S.C. 112, sixth paragraph, except as otherwise indicated in an Office action. Conversely, claim limitations in this application that do not use the word “means” (or “step”) are not being interpreted under 35 U.S.C. 112(f) or pre-AIA 35 U.S.C. 112, sixth paragraph, except as otherwise indicated in an Office action.
This application includes one or more claim limitations that do not use the word “means,” but are nonetheless being interpreted under 35 U.S.C. 112(f) or pre-AIA 35 U.S.C. 112, sixth paragraph, because the claim limitation(s) uses a generic placeholder that is coupled with functional language without reciting sufficient structure to perform the recited function and the generic placeholder is not preceded by a structural modifier. Such claim limitation(s) is/are:
"a beam condition input part" and "a signal output part" in claim 7.
The Examiner notes that:
regarding the limitation(s) of "a beam condition input part" and "a signal output part" in claim 7, is/are embodied as part of a processor, as per the as filed Specification ¶ in page 6 lines 14-17: A processor of the beamformer learning system 100 may include a beam condition input part 210 and a signal output part 220. The components of such a processor may be representations of different functions performed by the processor in accordance with a control instruction provided by a program code stored in the beamformer learning system.
Because this/these claim limitation(s) is/are being interpreted under 35 U.S.C. 112(f) or pre-AIA 35 U.S.C. 112, sixth paragraph, it/they is/are being interpreted to cover the corresponding structure described in the specification as performing the claimed function, and equivalents thereof.
If applicant does not intend to have this/these limitation(s) interpreted under 35 U.S.C. 112(f) or pre-AIA 35 U.S.C. 112, sixth paragraph, applicant may: (1) amend the claim limitation(s) to avoid it/them being interpreted under 35 U.S.C. 112(f) or pre-AIA 35 U.S.C. 112, sixth paragraph (e.g., by reciting sufficient structure to perform the claimed function); or (2) present a sufficient showing that the claim limitation(s) recite(s) sufficient structure to perform the claimed function so as to avoid it/them being interpreted under 35 U.S.C. 112(f) or pre-AIA 35 U.S.C. 112, sixth paragraph.
Claim Rejections - 35 USC § 112
The following is a quotation of 35 U.S.C. 112(b):
(b) CONCLUSION.—The specification shall conclude with one or more claims particularly pointing out and distinctly claiming the subject matter which the inventor or a joint inventor regards as the invention.
The following is a quotation of 35 U.S.C. 112 (pre-AIA ), second paragraph:
The specification shall conclude with one or more claims particularly pointing out and distinctly claiming the subject matter which the applicant regards as his invention.
Claims 1-7 rejected under 35 U.S.C. 112(b) or 35 U.S.C. 112 (pre-AIA ), second paragraph, as being indefinite for failing to particularly point out and distinctly claim the subject matter which the inventor or a joint inventor (or for applications subject to pre-AIA 35 U.S.C. 112, the applicant), regards as the invention.
Claim 1 recites the limitation "the method" in line 2 of the claim. There is insufficient antecedent basis for this limitation in the claim.
Hence, dependent claims 2-6 are also rejected.
Claims 1 and 7 recite the limitation "the direction of interest (DOI)" in line 4 of claims 1 and 7. There is insufficient antecedent basis for this limitation in the claim.
Hence, dependent claims 2-6 are also rejected.
Claim 4 recites the limitation "the source position" in line 3 of the claim. There is insufficient antecedent basis for this limitation in the claim.
Claim 5 recites the limitation "the direction of arrival (DOA)" in line 2 of the claim. There is insufficient antecedent basis for this limitation in the claim.
Claim Rejections - 35 USC § 101
35 U.S.C. 101 reads as follows:
Whoever invents or discovers any new and useful process, machine, manufacture, or composition of matter, or any new and useful improvement thereof, may obtain a patent therefor, subject to the conditions and requirements of this title.
Claim(s) 1-7 rejected under 35 U.S.C. 101 because the claimed invention is directed to an abstract idea without significantly more. More specifically directed to the abstract idea grouping of: mental process and/or mathematical concepts.
The independent claim(s) 1 and 7 recite(s):
Claim 1. A supervised learning method for spatial filtering of speech, performed by a beamformer learning system, the method comprising:
receiving, as input into a neural network-based beamformer model, a multi-channel speech signal incident on a microarray in a reverberant environment and a beam condition representing the direction of interest (DOI); and
outputting a desired signal corresponding to the beam condition from the multi-channel speech signal by using the neural network-based beamformer model,
wherein the neural network-based beamformer model is trained to extract a speech signal with azimuth and elevation angles that are set for the beam condition, by using training data.
Claim 7. A beamformer learning system comprising:
a beam condition input part that receives, as input into a neural network-based beamformer model, a multi-channel speech signal incident on a microarray in a reverberant environment and a beam condition representing the direction of interest (DOI); and
a signal output part that outputs a desired signal corresponding to the beam condition from the multi-channel speech signal by using the neural network-based beamformer model,
wherein the neural network-based beamformer model is trained to extract a speech signal with azimuth and elevation angles that are set for the beam condition, by using training data.
This reads on a human (e.g., mentally and/or using pen and paper):
Receiving data (e.g., signal in a noisy environment) and a predefined condition in order to follow a predetermined set of rules;
Filtering the received data based on the predefined condition by using the predetermined set of rules;
Wherein the predetermined set of rules is updated/defined/re-defined to extract values (i.e., azimuth/angles).
This judicial exception is not integrated into a practical application because for example: claim 1 recites “beamforming learning system”, “a neural network-based beamformer model”, “a multi-channel speech signal incident on a microarray” while claim 7 further recites “a beam condition input part” and “a signal output part”. As an example, in page 12, lines 1-8 of the as filed specification, it is disclosed: The apparatus described above may be implemented as a hardware component, a software component, and/or a combination of the hardware component and the software component. For example, devices and components described in the embodiments may be implemented using, for example, one or more general purpose or special purpose computers, such as a processor, a controller, an arithmetic logic unit (ALU), a digital signal processor, a microcomputer, a field programmable gate array (FPGA), a programmable logic unit (PLU), a microprocessor, or any other device capable of executing and responding to instructions.. Therefore, a general-purpose computer or computing device is described and mainly used as an application thereof. Accordingly, these additional elements do not integrate the abstract idea into a practical idea because it does not impose any meaningful limits on practicing the abstract idea.
The claim(s) does/do not include additional elements that are sufficient to amount to significantly more than the judicial exception. As discussed above with respect to the integration of the abstract idea into a practical application, the additional elements of using a computer is listed as a general computing device as noted. The claim is not patent eligible.
With respect to claims 2, the claim(s) recite:
Claim 2. The supervised learning method of claim 1,
wherein spatial gain functions are configured to define a desired signal determined according to the beam condition,
wherein the spatial gain functions include a hard gain function and a soft gain function.
This reads on a human (e.g., mentally and/or using pen and paper):
Defining gain functions (e.g., mathematical concepts).
No additional limitations are present.
With respect to claims 3, the claim(s) recite:
Claim 3. The supervised learning method of claim 1,
wherein the receiving comprises generating training data to train the neural network-based beamformer model with a spatial filter using a supervised learning method.
This reads on a human (e.g., mentally and/or using pen and paper):
Updating/defining/re-defining predetermined set of rules (e.g., mathematical concepts).
No additional limitations are present.
With respect to claims 4, the claim(s) recite:
Claim 4. The supervised learning method of claim 3,
wherein the receiving comprises determining a beam condition for look direction and beamwidth through early reflections multiplied by spatial gain and multiple different combinations for the source position and DOI parameters.
This reads on a human (e.g., mentally and/or using pen and paper):
Extracting or determining values/parameters (e.g., mathematical concepts).
No additional limitations are present.
With respect to claims 5, the claim(s) recite:
Claim 5. The supervised learning method of claim 1,
wherein the receiving comprises obtaining single-path propagations of the early reflections by using the direction-of-arrival (DOA) of a direct path in multiple paths and an image method.
This reads on a human (e.g., mentally and/or using pen and paper):
Extracting or determining values/parameters (e.g., mathematical concepts).
No additional limitations are present.
With respect to claims 6, the claim(s) recite:
Claim 6. The supervised learning method of claim 1,
wherein the receiving comprises defining DOI information for specifying direction information and a range of interest in a three-dimensional space, and converting the defined DOI information into a beam condition vector.
This reads on a human (e.g., mentally and/or using pen and paper):
Extracting or determining values/parameters (e.g., mathematical concepts).
No additional limitations are present.
Claim Rejections - 35 USC § 103
The following is a quotation of 35 U.S.C. 103 which forms the basis for all obviousness rejections set forth in this Office action:
A patent for a claimed invention may not be obtained, notwithstanding that the claimed invention is not identically disclosed as set forth in section 102, if the differences between the claimed invention and the prior art are such that the claimed invention as a whole would have been obvious before the effective filing date of the claimed invention to a person having ordinary skill in the art to which the claimed invention pertains. Patentability shall not be negated by the manner in which the invention was made.
The factual inquiries for establishing a background for determining obviousness under 35 U.S.C. 103 are summarized as follows:
1. Determining the scope and contents of the prior art.
2. Ascertaining the differences between the prior art and the claims at issue.
3. Resolving the level of ordinary skill in the pertinent art.
4. Considering objective evidence present in the application indicating obviousness or nonobviousness.
This application currently names joint inventors. In considering patentability of the claims the examiner presumes that the subject matter of the various claims was commonly owned as of the effective filing date of the claimed invention(s) absent any evidence to the contrary. Applicant is advised of the obligation under 37 CFR 1.56 to point out the inventor and effective filing dates of each claim that was not commonly owned as of the effective filing date of the later invention in order for the examiner to consider the applicability of 35 U.S.C. 102(b)(2)(C) for any potential 35 U.S.C. 102(a)(2) prior art against the later invention.
Claims 1 and 6 is/are rejected under 35 U.S.C. 103 as being unpatentable over Xu et al. (Xu, Yong, et al. "Neural spatio-temporal beamformer for target speech separation." arXiv preprint arXiv:2005.03889 (2020).) and further in view of Ebenezer (US 20180102135 A1), and Zegrar et al. (US 20230421412 A1).
As to independent claim 1, Xu et al. teaches:
Claim 1. A (see ¶ 4 of section 1. Introduction: “In this work, we propose a neural spatio-temporal beamforming approach, named multi-tap MVDR beamformer with complex-valued masks, for speech separation and enhancement to simultaneously obtain high ASR accuracy and PESQ score…”), the method comprising:
receiving, as input into a neural network-based beamformer model, a multi-channel speech signal incident on a microarray in a reverberant environment and a beam condition representing the direction of interest (DOI) (see Fig. 1 (multi-channel mixture, 15-element linear microphone array is aligned with the camera, target DOA, and Multi-tap MVDR ) and ¶ 1 and 3 of section 2. Neural Spatio-Temporal Beamformer: Multi-tap MVDR with Complex Mask: ¶ 1: “2.1. Spatial filtering: MVDR beamformer MVDR is a widely used beamformer for ASR [18]. It minimizes the power of the noise (interfering speech + additive noise) while ensuring that the signal at the desired direction is not distorted.” and ¶ 3: “2.2. Neural spatial filtering: Mask based MVDR The idea behind the mask based approach for covariance matrix estimation is that we may more accurately estimate the target speech, and thus the covariance matrix, given a NN-based mask estimator (will discuss in Sec. 3.1).The most commonly used mask for the mask-based beamforming [18] is ideal ratio mask (IRM) [32] or sigmoid mask.…” and ¶ 2 and 4 of section 3. Experimental Setup and Baselines: ¶ 2: “As shown in Fig. 1, we use the direction of arrival (DOA) of the target speaker and the speaker-dependent lip sequence for informing the dilated convolutional neural networks (CNNs) to extract the target speech from the multi-talker mixture.” ¶ 4: Audio encoder: The audio input includes the speaker independent features (e.g., log-power spectra (LPS) and interaural phase difference (IPD) [13]) and speaker-dependent feature (e.g., directional feature dpq [41, 15]). As shown in Fig. 1, the 15-element non-uniform linear microphone array [13] is co-located with the 180wide-angle camera. The location of the target speaker’s face in the whole camera view can provide a rough DOA estimation of the target speaker…”); and
outputting a desired signal corresponding to the beam condition from the multi-channel speech signal by using the neural network-based beamformer model (see Fig. 1 (multi-channel mixture, 15-element linear microphone array is aligned with the camera, target DOA, Multi-tap MVDR, and predicted target waveform) and ¶ 7-9 of section 2. Neural Spatio-Temporal Beamformer: Multi-tap MVDR with Complex Mask: ¶ 7: “Inspired by the single channel multi-frame MVDR [26,27, 29] which utilizes the inter-frame correlation, we propose a multi-tap MVDR for the multi-channel neural beamforming to achieve distortionless speech and low residual noise simultaneously...” ¶ 8-9: “…The enhanced speech of the multi-tap MVDR can be obtained as, ^Spt; fq wH pfqYpt; fq (11) The beamformed spectrum is converted to the time-domain waveform via iSTFT…” and further ¶ 4 of section 3. Experimental Setup and Baselines: “…The location of the target speaker’s face in the whole camera view can provide a rough DOA estimation of the target speaker. Chen et al [41] proposed a location guided directional feature (DF) dpq to extract the target speech from the specific DOA…”),
wherein the neural network-based beamformer model is trained to extract a speech signal (see ¶ 7-9 of section 2. Neural Spatio-Temporal Beamformer: Multi-tap MVDR with Complex Mask: ¶ 7: “Inspired by the single channel multi-frame MVDR [26,27, 29] which utilizes the inter-frame correlation, we propose a multi-tap MVDR for the multi-channel neural beamforming to achieve distortionless speech and low residual noise simultaneously...” and further ¶ 6 of section 3. Experimental Setup and Baselines: “The multi-modal network is trained in a chunk-wise mode with chunk size 4 seconds, using Adam optimizer with early stopping. Initial learning rate is set to 1e-3. The L-tap in the multi-tap MVDR is set to 3 empirically. Pytorch 1.1.0 was used…”).
PNG
media_image1.png
404
898
media_image1.png
Greyscale
However, Xu et al. does not explicitly teach, but Ebenezer does teach:
supervised learning method for spatial filtering of speech (see ¶ [0020]: “… An output power from each beamformer may be computed, and a beamformer 33 having a maximum output power may be switched to an output path 34 by a beam selector 35. Switching of beam selector 35 may be constrained by a voice activity detector 31 having an impulsive noise detector 32 such that the output power is measured by beam selector 35 only when speech is detected, thus preventing beam selector 35 from rapidly switching between multiple beamformers 33 by responding to spatially non-stationary background impulsive noises.” and ¶ [0034]: “…Fusion logic 60 may apply a supervised learning algorithm such as, for example, a support vector machine (SVM) to determine a non-linear function that optimally separates speech and impulse noise in a four-dimensional feature space, custom-character.sup.4, each dimension of the feature space corresponding to one of the foregoing parameters (e.g., harmonicity, harmonic product spectrum flatness measure, spectral flatness measure, and spectral flatness measure swing). For example, FIGS. 6A-6F show the distribution of pair-wise statistics and the decision boundary generated by the SVM (e.g., linear kernel) when only two of the four statistics are used…”)
Xu et al. and Ebenezer are considered to be analogous to the claimed invention because they are in the same field of endeavor in communications / signal processing. Therefore, it would have been obvious to someone of ordinary skill in the art before the effective filing date of the claimed invention to have modified Xu et al. to incorporate the teachings of Ebenezer of supervised learning method for spatial filtering of speech which provides the benefit of multi-channel signal enhancement ([0034] of Ebenezer).
However, Xu et al. in combination with Ebenezer do not explicitly teach, but Zegrar et al. does teach:
wherein the neural network-based (i.e., as taught by Xu et al.) beamformer model is trained to extract a speech signal with azimuth and elevation angles that are set for the beam condition, by using training data (see ¶ [0091]: “Here, θ.sub.AoA.sup.R is an elevation AoA at the configurable surface, φ.sub.AoA.sup.R is an azimuth AoA at the configurable surface. As mentioned above, θ.sub.AoD.sup.R is an elevation AoD at the configurable surface, φ.sub.AoD.sup.R is an azimuth AoD at the configurable surface, and these may be obtained by the beamforming training…” and ¶ [0181]: “… By performing the beamforming search, i) trained reflection coefficients of the configurable surface (120), and ii) an angle of arrival (AoA) of the signals at the receiving device (135) are obtained. ”)
Xu et al., Ebenezer and Zegrar et al. are considered to be analogous to the claimed invention because they are in the same field of endeavor in communications / signal processing. Therefore, it would have been obvious to someone of ordinary skill in the art before the effective filing date of the claimed invention to have modified Xu et al. in combination with Ebenezer to incorporate the teachings of Zegrar et al. of wherein the beamformer model is trained to extract a speech signal with azimuth and elevation angles that are set for the beam condition, by using training data which provides the benefit of enhancing received signal power ([0034] of Zegrar et al.).
Regarding claim 6, Xu et al. in combination with Ebenezer and Zegrar et al. teaches the limitations as in claim 1, above.
Zu et al. further teaches:
Claim 6. The supervised learning method of claim 1,
wherein the receiving comprises defining DOI information for specifying direction information and a range of interest in a three-dimensional space, and converting the defined DOI information into a beam condition vector (see Fig. 1 (multi-channel mixture, video captured by 180 deg wide-angle camera, 15-element linear microphone array is aligned with the camera, target DOA, Multi-tap MVDR, and predicted target waveform) and ¶ 7-9 of section 2. Neural Spatio-Temporal Beamformer: Multi-tap MVDR with Complex Mask: ¶ 7: “Inspired by the single channel multi-frame MVDR [26,27, 29] which utilizes the inter-frame correlation, we propose a multi-tap MVDR for the multi-channel neural beamforming to achieve distortionless speech and low residual noise simultaneously...” ¶ 8-9: “…The enhanced speech of the multi-tap MVDR can be obtained as, ^Spt; fq wH pfqYpt; fq (11) The beamformed spectrum is converted to the time-domain waveform via iSTFT…” and further ¶ 4 of section 3. Experimental Setup and Baselines: “…The location of the target speaker’s face in the whole camera view can provide a rough DOA estimation of the target speaker. Chen et al [41] proposed a location guided directional feature (DF) dpq to extract the target speech from the specific DOA…”and further ¶ 4 of section 3. Experimental Setup and Baselines: “Audio encoder: The audio input includes the speaker independent features (e.g., log-power spectra (LPS) and interaural phase difference (IPD) [13]) and speaker-dependent feature (e.g., directional feature dpq [41, 15]). As shown in Fig. 1, the 15-element non-uniform linear microphone array [13] is co-located with the 180wide-angle camera. The location of the target speaker’s face in the whole camera view can provide a rough DOA estimation of the target speaker. Chen et al [41] proposed a location guided directional feature (DF) dpq to extract the target speech from the specific DOA.”).
Claim 2 is/are rejected under 35 U.S.C. 103 as being unpatentable over Xu et al. (Xu, Yong, et al. "Neural spatio-temporal beamformer for target speech separation." arXiv preprint arXiv:2005.03889 (2020).) and further in view of Ebenezer (US 20180102135 A1), and Zegrar et al. (US 20230421412 A1) as applied to claim 1 above, and further in view of Kokkinakis et al. (US 20210166713 A1).
Regarding claim 2, Xu et al. in combination with Ebenezer and Zegrar et al. teaches the limitations as in claim 1, above.
However, Xu et al. in combination with Ebenezer and Zegrar et al. do not explicitly teach, but Kokkinakis et al. does teach:
Claim 2. The supervised learning method of claim 1,
wherein spatial gain functions are configured to define a desired signal determined according to the beam condition (see ¶ [0040]: “The gain estimation stages discussed herein do not require access to a theoretical clean or an uncorrupted signal, and therefore the present approach is ‘blind’ and generalizable to any acoustical environment. The statistical parameters necessary to form either the hard or soft decision masks can be easily adapted based on information extracted exclusively from the microphone signal outputs. Algorithms can be easily integrated in existing audio processors equipped with two spaced-apart external microphones and can operate in parallel or in conjunction with a beamforming module to enhance the acoustic input. Such embodiments provide a robust technique for suppression of room reverberation inherent in the signals recorded by two spatially separated microphones, and also can provide adequate suppression of background noise from a number of interfering speakers.”),
wherein the spatial gain functions include a hard gain function and a soft gain function (see ¶ [0040] citation as in limitation above: “hard/soft decision masks”).
Xu et al., Ebenezer, Zegrar et al., and Kokkinakis et al. are considered to be analogous to the claimed invention because they are in the same field of endeavor in communications / signal processing. Therefore, it would have been obvious to someone of ordinary skill in the art before the effective filing date of the claimed invention to have modified Xu et al. in combination with Ebenezer and Zegrar et al. to incorporate the teachings of Kokkinakis et al. of wherein spatial gain functions are configured to define a desired signal determined according to the beam condition, wherein the spatial gain functions include a hard gain function and a soft gain function which provides the benefit of providing a robust technique for suppression of room reverberation inherent in the signals recorded by two spatially separated microphones, and also providing adequate suppression of background noise from a number of interfering speakers([0040] of Kokkinakis et al.).
Claim 3 is/are rejected under 35 U.S.C. 103 as being unpatentable over Xu et al. (Xu, Yong, et al. "Neural spatio-temporal beamformer for target speech separation." arXiv preprint arXiv:2005.03889 (2020).) and further in view of Ebenezer (US 20180102135 A1), and Zegrar et al. (US 20230421412 A1) as applied to claim 1 above, and further in view of Orr et al. (US 20220308166 A1).
Regarding claim 3, Xu et al. in combination with Ebenezer and Zegrar et al. teaches the limitations as in claim 1, above.
However, Xu et al. in combination with Ebenezer and Zegrar et al. do not explicitly teach, but Orr et al. does teach:
Claim 3. The supervised learning method of claim 1,
wherein the receiving comprises generating training data to train the neural network-based beamformer model with a spatial filter using a supervised learning method (see ¶ [0012 and 0054]: “[0012] Multiple snapshots of a spatial covariance matrix were used with a Convolutional Neural Network (CNN) and a 1D antenna array with simulated data for Direction of Arrival (DOA) estimation and super-resolution. A single snapshot of a spatial covariance matrix was used with a fully connected model for DOA estimation and super resolution of a 2D antenna array with simulated data and a 1D antenna array with both simulation and real-world data where the targets were corner reflectors… [0054] Accordingly, some embodiments of the invention enable real time performance of a DNN based system for coherent RADAR beamforming and super resolution that can be utilized for automotive applications in real-world, urban and highway scenarios. More specifically, some embodiments use an auto-encoder trained in a self-supervised method with a diluted RADAR array and used to reconstruct the amplitude and phase of missing receiving channels. To enforce coherence during the reconstruction process, a novel loss function which operates on multiple data representation spaces may be utilized.”)
Xu et al., Ebenezer, Zegrar et al., and Orr et al. are considered to be analogous to the claimed invention because they are in the same field of endeavor in signal processing. Therefore, it would have been obvious to someone of ordinary skill in the art before the effective filing date of the claimed invention to have modified Xu et al. in combination with Ebenezer and Zegrar et al. to incorporate the teachings of Orr et al. of wherein the receiving comprises generating training data to train the neural network-based beamformer model with a spatial filter using a supervised learning method which provides the benefit of improving the resolution of any system ([0057] of Orr et al.).
Claim 4 is/are rejected under 35 U.S.C. 103 as being unpatentable over Xu et al. (Xu, Yong, et al. "Neural spatio-temporal beamformer for target speech separation." arXiv preprint arXiv:2005.03889 (2020).) and further in view of Ebenezer (US 20180102135 A1), and Zegrar et al. (US 20230421412 A1) and Orr et al. (US 20220308166 A1) as applied to claim 3 above, and further in view of Reynolds et al. (US 11555916 B2).
Regarding claim 4, Xu et al. in combination with Ebenezer and Zegrar et al. and Orr et al. teaches the limitations as in claim 3, above.
However, Xu et al. in combination with Ebenezer and Zegrar et al. and Orr et al. do not explicitly teach, but Reynolds et al. does teach:
Claim 4. The supervised learning method of claim 3,
wherein the receiving comprises determining a beam condition for look direction and beamwidth through early reflections multiplied by spatial gain and multiple different combinations for the source position and DOI parameters (see ¶ Col. 1, lines 45-50: “(3) In active mmW and/or microwave imaging, the region of interest (ROI) in a scene is illuminated by one or more transmitters, while the scattered energy from the scene is spatially sampled by one or more antennas and receivers. ” and ¶ Col. 14, line 53 – Col. 16, line 11: “(42) The multiple beams of orthogonal signals may be orthogonal in time, frequency, and/or code. For example, time division multiplexing of the multiple beams of orthogonal signals may be provided by controlling the antennas 106 (e.g., a plurality of radiating elements) to emit each of the multiple beams of orthogonal signals at a different time. Frequency division multiplexing may be provided by controlling the antennas 106 (e.g., a plurality of radiating elements) to emit each of the multiple beams of orthogonal signals at a different frequency, or with a modulation scheme such as orthogonal frequency division multiplexing (OFDM). Code division multiplexing may be provided by controlling the antennas 106 (e.g., a plurality of radiating elements) to emit each of the multiple beams of orthogonal signals utilizing a different code. Such a code may include an orthogonal code such as a Barker code, a pseudonoise (PN) code, or another sequence of modulation symbols that are orthogonally disposed. The controller 120 (e.g., using RF transceiver control) may provide the described control over the beams and over the orthogonal signals. (43) The RF transceiver 108 may receive scattered signals resulting from the scattering of the multiple beams of orthogonal signals from the scene, and the reconstruction system 124 and/or an imaging device may utilize the energy from the scattered signals to generate image data of the scene. In some examples, the reconstruction system 124 and/or the imaging device may utilize partitioned inverse techniques described herein to provide the image data. The partitioned inverse techniques may include utilizing a measurement matrix partitioned in accordance with the multiple beams of orthogonal signals. (44) Examples described herein accordingly provide image reconstruction (e.g., SAR image reconstruction) based on scattered signals from a scene responsive to interrogation by one or more interrogation signals, Images described herein may accordingly be considered representations of the complex radar reflectivity of scenes (e.g., scene 102 of FIG. 1). One or more antennas (e.g., antennas 106) may transmit interrogation signal(s) and/or receive scattered signals (e.g., backscattered energy) due to the illuminated point scatterers comprising the scene. For example, the scene may be considered to be (e.g., may be modeled as) a collection of point scatterers. This yields a position-dependent scene transfer function. Interrogation signals described herein may be single frequency, narrow-band, wideband, or ultra-wideband waveforms in some examples, and the ground-range resolution may be inversely proportional to the bandwidth. The cross-range (azimuth) resolution may be inversely proportional to the length of the synthetic aperture. Therefore, with relative motion between the scene and the interrogating antennas (e.g., as the antennas or scene move relative to each other transfer functions of the scene may be measured at different positions to create a large effective aperture. For example, at antenna positions m∈{1, . . . , M}, the sampled frequency response vector y.sub.m may be measured. The antenna positions may refer, for example, to positions of the antennas 106 of FIG. 1, and the sampled frequency response vector y.sub.m may be measured using, for example receivers in RF transceiver 108 of FIG. 1. Assuming a scene consisting of N point scatterers, the received signal y.sub.m,l, at the l-th frequency sample, where l∈{1, . . . , L}, of the measurement vector y.sub.m may be written as: (Equation 1) where is the complex reflection coefficient (e.g., the reflectivity value) of the n-th point scatterer, a.sub.m,l,n is the propagation channel including path loss and antenna response, cot is the angular frequency, is the round-trip time delay between antenna position m and the n-th point scatter, is measurement noise, and j=√−1. Concatenating all M measurements (at each of M positions) gives y=Hx+v, where y is a vector of measurements (e.g., measurements of scattered signals at M positions) y=[y.sub.1.sup.T, . . . , y.sub.M.sup.T].sup.T, v is measurement noise, and x is a vector of reflectivity values of modeled point scatterers making up the scene, x=[x.sub.1, . . . , x.sub.N].sup.T is the reflectivity vector of the point scatterers, and (.).sup.T is the transpose operator. The measurement matrix H provides a dependence between the reflectivity values of the point scatterers and each measurement. The elements of the measurement matrix are defined as [H]m(L−1)+l, n=a.sub.m,l.ne.sup.−jω.sup.l.sup.t.sup.m,n. (Equation 2) where in is the index of each position of the antennas, 1 is the index of the frequency component of the scattered signal (e.g., each frequency component index through L), a.sub.m,l,n is the propagation channel including path loss and antenna response, qui is the angular frequency, t.sub.m,n is the round-trip time delay between antenna position m and the n-th point scatter, v.sub.m,l is measurement noise, and j=√−1.”).
Xu et al., Ebenezer, Zegrar et al., and Reynolds et al. are considered to be analogous to the claimed invention because they are in the same field of endeavor in signal processing. Therefore, it would have been obvious to someone of ordinary skill in the art before the effective filing date of the claimed invention to have modified Xu et al. in combination with Ebenezer and Zegrar et al. to incorporate the teachings of Reynolds et al. of wherein the receiving comprises determining a beam condition for look direction and beamwidth through early reflections multiplied by spatial gain and multiple different combinations for the source position and DOI parameters which provides the benefit of improving resolution while maintaining good SNR (Col. 21, line 44 of Reynolds et al.).
Claim 5 is/are rejected under 35 U.S.C. 103 as being unpatentable over Xu et al. (Xu, Yong, et al. "Neural spatio-temporal beamformer for target speech separation." arXiv preprint arXiv:2005.03889 (2020).) and further in view of Ebenezer (US 20180102135 A1), and Zegrar et al. (US 20230421412 A1) as applied to claim 1 above, and further in view of Cai et al. (US 20230117339 A1).
Regarding claim 5, Xu et al. in combination with Ebenezer and Zegrar et al. teaches the limitations as in claim 1, above.
However, Xu et al. in combination with Ebenezer and Zegrar et al. do not explicitly teach, but Cai et al. does teach:
Claim 5. The supervised learning method of claim 1,
wherein the receiving comprises obtaining single-path propagations of the early reflections by using the direction-of-arrival (DOA) of a direct path in multiple paths and an image method (see ¶ [0040 and 0062]: “[0040] Using the MIMO antenna 114, some implementations of the radar system 104 can form beams that are steered or un-steered, and wide or narrow. The steering and shaping can be achieved through analog beamforming or digital beamforming… [0062] The electromagnetic simulator 134 considers five propagation paths with multipath effects as illustrated in FIGS. 2-4. FIG. 2-4(a) illustrates a path 244 reflecting directly back to the transceiver 236 if the path is not blocked by other facets. FIGS. 2-4 illustrates a path 246 that reflects back to the transceiver image 248. This is equivalent to a path that reflects off the ground and to the transceiver 236. In FIG. 2-4(c), a path 250 reverses back over an incident path. This path 250 is considered if it is not the specular reflection direction of the target. FIGS. 2-4 and (e) are the reciprocals of the paths 244 and 246 illustrated in FIGS. 2-4(a) and (b), respectively. That is, the paths 252 and 254 have the same electromagnetic responses as the paths 244 and 246, respectively, but are in opposite directions of departure and arrival (e.g., inverse paths). By evaluating these extra five paths, the electromagnetic simulator 134 can more accurately simulate the real world.”).
Xu et al., Ebenezer, Zegrar et al., and Cai et al. are considered to be analogous to the claimed invention because they are in the same field of endeavor in signal processing. Therefore, it would have been obvious to someone of ordinary skill in the art before the effective filing date of the claimed invention to have modified Xu et al. in combination with Ebenezer and Zegrar et al. to incorporate the teachings of Cai et al. of wherein the receiving comprises obtaining single-path propagations of the early reflections by using the direction-of-arrival (DOA) of a direct path in multiple paths and an image method which provides the benefit of improving horizontal multipath effects ([0030] of Cai et al.).
Claim 7 is/are rejected under 35 U.S.C. 103 as being unpatentable over Xu et al. (Xu, Yong, et al. "Neural spatio-temporal beamformer for target speech separation." arXiv preprint arXiv:2005.03889 (2020).) and further in view of Zegrar et al. (US 20230421412 A1).
As to independent claim 7, Xu et al. teaches:
Claim 7. A beamformer learning system (see ¶ 4 of section 1. Introduction citation as in claim 1, above and further ¶ 1 of. Section 4. Results and Discussions: “The PESQ and ASR word error rate (WER) results are shown in Table 1 to compare among purely network-based systems and several jointly trained MVDR systems…”) comprising:
a beam condition input part (see Fig. 1 (multi-channel mixture, 15-element linear microphone array is aligned with the camera, target DOA, and Multi-tap MVDR ) and ¶ 2 and 4 of section 3. Experimental Setup and Baselines: ¶ 2: “As shown in Fig. 1, we use the direction of arrival (DOA) of the target speaker and the speaker-dependent lip sequence for informing the dilated convolutional neural networks (CNNs) to extract the target speech from the multi-talker mixture.” ¶ 4: Audio encoder: The audio input includes the speaker independent features (e.g., log-power spectra (LPS) and interaural phase difference (IPD) [13]) and speaker-dependent feature (e.g., directional feature dpq [41, 15]). As shown in Fig. 1, the 15-element non-uniform linear microphone array [13] is co-located with the 180 wide-angle camera. The location of the target speaker’s face in the whole camera view can provide a rough DOA estimation of the target speaker…”) that receives, as input into a neural network-based beamformer model, a multi-channel speech signal incident on a microarray in a reverberant environment and a beam condition representing the direction of interest (DOI) (see Fig. 1, ¶ 1 and 3 of section 2. Neural Spatio-Temporal Beamformer: Multi-tap MVDR with Complex Mask and ¶ 2 and 4 of section 3. Experimental Setup and Baselines citations as in claim 1, above.); and
a signal output part (see Fig. 1 (multi-channel mixture, 15-element linear microphone array is aligned with the camera, target DOA, Multi-tap MVDR, and predicted target waveform)) that outputs a desired signal corresponding to the beam condition from the multi-channel speech signal by using the neural network-based beamformer model (see Fig. 1, ¶ 7-9 of section 2. Neural Spatio-Temporal Beamformer: Multi-tap MVDR with Complex Mask and ¶ 4 of section 3. Experimental Setup and Baselines citations as in claim 1, above.),
wherein the neural network-based beamformer model is trained to extract a speech signal (see ¶ 7-9 of section 2. Neural Spatio-Temporal Beamformer: Multi-tap MVDR with Complex Mask and ¶ 6 of section 3. Experimental Setup and Baselines citations as in claim 1, above.).
However, Xu et al. does not explicitly teach, but Zegrar et al. does teach:
wherein the neural network-based (i.e., as taught by Xu et al.) beamformer model is trained to extract a speech signal with azimuth and elevation angles that are set for the beam condition, by using training data (see ¶ [0091 and 0181] citations as in claim 1, above.)
Xu et al. and Zegrar et al. are considered to be analogous to the claimed invention because they are in the same field of endeavor in communications / signal processing. Therefore, it would have been obvious to someone of ordinary skill in the art before the effective filing date of the claimed invention to have modified Xu et al. to incorporate the teachings of Zegrar et al. of wherein the beamformer model is trained to extract a speech signal with azimuth and elevation angles that are set for the beam condition, by using training data which provides the benefit of enhancing received signal power ([0034] of Zegrar et al.).
Conclusion
The prior art made of record and not relied upon is considered pertinent to applicant's disclosure.
Pertinent to claims 1-7:
US-12300261-B1
US-20200213726-A1
US-20230095526-A1
US-20200402526-A1
Li et al. (“Deep neural network-based generalized sidelobe canceller for dual-channel far-field speech recognition”, (2021)) (Year: 2021)
Liu et al. (“Causal Deep CASA for Monaural Talker-Independent Speaker Separation”, (2020)) (Year: 2020)
Any inquiry concerning this communication or earlier communications from the examiner should be directed to Keisha Y Castillo-Torres whose telephone number is (571)272-3975. The examiner can normally be reached Monday - Friday, 9:00 am - 4:00 pm (EST).
Examiner interviews are available via telephone, in-person, and video conferencing using a USPTO supplied web-based collaboration tool. To schedule an interview, applicant is encouraged to use the USPTO Automated Interview Request (AIR) at http://www.uspto.gov/interviewpractice.
If attempts to reach the examiner by telephone are unsuccessful, the examiner’s supervisor, Pierre-Louis Desir can be reached at (571)272-7799. The fax phone number for the organization where this application or proceeding is assigned is 571-273-8300.
Information regarding the status of published or unpublished applications may be obtained from Patent Center. Unpublished application information in Patent Center is available to registered users. To file and manage patent submissions in Patent Center, visit: https://patentcenter.uspto.gov. Visit https://www.uspto.gov/patents/apply/patent-center for more information about Patent Center and https://www.uspto.gov/patents/docx for information about filing in DOCX format. For additional questions, contact the Electronic Business Center (EBC) at 866-217-9197 (toll-free). If you would like assistance from a USPTO Customer Service Representative, call 800-786-9199 (IN USA OR CANADA) or 571-272-1000.
Keisha Y. Castillo-Torres
Examiner
Art Unit 2659
/Keisha Y. Castillo-Torres/Examiner, Art Unit 2659
Prosecution Insights