DETAILED ACTION
The present application, filed on or after March 16, 2013, is being examined under the first inventor to file provisions of the AIA .
In the response to this office action, the Examiner respectfully requests that support be shown for language added to any original claims on amendment and any new claims. That is, indicate support for newly added claim language by specifically pointing to page(s) and line numbers in the specification and/or drawing figure(s). This will assist the Examiner in prosecuting this application.
Claim Objections
Claims 1-20 are objected to because of the following informalities:
Claim 1 recited “… detecting speech from …”, “for extracting a speech”, and “extracting residual speech masked by …”, which should be --… detecting a speech from …--, --for extracting [[a]]the speech--, and --extracting a residual speech masked by …--, respectively. Claims 2-13 are objected due to the dependencies to claim 1.
Claim 14 is objected for the at least similar reasons as described in claim 1 above since claim 14 recited similar deficient features as recited in claim 1.
Claim 15 is objected for the at least similar reasons as described in claim 1 above since claim 15 recited similar deficient features as recited in claim 1. Claims 16-20 are objected due to the dependencies to claim 15.
Appropriate correction is required.
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.
Claims 1-20 are rejected under 35 U.S.C. 112(b) as being indefinite for failing to particularly point out and distinctly claim the subject matter which applicant regards as the invention.
Claim 1 recites “a first component” of “each masking block” of “a series of masking blocks”, “that generates a first mask”, i.e., “a first component of … a series of masking blocks” produces or generates “a first mask” for each of “masking block” and claim 1 further recites “… the first masks produced by the series of masking blocks …”, i.e., “the first masks” are “produced by the series of masking blocks”, which is confusing because it is unclear whether each “first component” of “series of masking blocks” produces/generates “the first masks” or “series of masking blocks” produce “the first masks” and thus, renders claim indefinite. This is because claim 1 recited a narrow limitation that it is “first component” of the “series of masking blocks” to generate “first mask(s)”, and then recited a broad limitation that it is “series of masking blocks” to generate the “first mask(s)”, and thus, it is unclear which one, “first component(s)” of the “series of masking blocks” or the “series of masking blocks”, generates the “first masks”. Claim 1 further recites “transmitting information related the first masks … to a device” and claim 1 further recites “masking values” included in each “first mask” by claiming “each mask of the first mask and the second mask including mask values”, “residual speech masked by the first mask”, estimated “an amount of speech present for each frame …” by “mask values” in claimed limitation “mask values estimating an amount of speech present for each frame …”, i.e., “masking values”, “residual speech”, “an amount of speech present” above are of “related the first masks”, and it is confusing because it is unclear whether the claimed “information related the first masks” is referred to one, two, some of, or all of “masking values”, “residual speech”, and “an amount of speech present” and thus, further renders claim indefinite. Claims 2-13 rejected due to the dependencies to claim 1.
Claim 14 is rejected for the at least similar reasons described in claim 1 above since claim 14 recited the similar deficient features as recited in claim 1 as discussed above.
Claim 15 recites “the digital model comprising a masking block comprising a first series of CNN blocks that generates a first mask … and a GRU block that generates a second mask for extracting residual speech masked by the first mask” and then recites “transmitting information related the first mask and the second mask”, which is confusing because it is unclear whether “information” herein is referred to “a first series of CNN blocks” that is “related the first mask and the second mask” or referred to extracted “residual speech” that is also “related the first mask and the second mask” and thus, renders claim indefinite. Claims 16-20 are rejected due to the dependencies to claim 15.
It is recommended to amend claims 1, 14, 15 and clearly recite what is to be “transmitted” and/or what “information” is about the claimed terms recited previous in claims 1, 14, 15.
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 of this title, 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.
Claims 1-4, 8-9, 13-14 are rejected under 35 U.S.C. 103 as being unpatentable over Kupryjanow et al. (US 20210012767 A1, hereinafter Kupryjanow) and in view of reference Shin et al. (US 20230037939 A1, hereinafter Shin).
Claim 1: Kupryjanow teaches a computer-implemented method of mitigating audio artifacts (title and abstract, ln 1-18, method steps in fig. 2, separating the speech from the dynamic noise and extracting the speech, para 22), comprising:
receiving, by a processor (implemented by a DSP in fig. 8, para 68), audio data as a joint time-frequency representation over a plurality of frames and a plurality of frequency bands (through STFT 102 to form Xk,n, k-frequency bin and n-frame, i.e., time-frequency domain, and at least used for training, para 38-43);
executing, by the processor, a digital model (including cascaded components 204 as part of temporal convolution network TCN in fig. 2) for detecting speech (extracting speech and remove/suppress dynamic noises from an audio stream during voice calls, para 14) from a feature vector of the audio data (magnitude spectrum and phase spectrum in fig. 2 or output layer norm 208 and normalized distribution of layers for smoother gradients, and better generalization accuracy, para 26),
the digital model comprising a series of masking blocks (a pair of 1-D dilated convolution blocks 204s and the pairs within the 204 stack repeated n times, para 25 or the stack repeated N times, e.g., eight stacked layers and three repeats, para 27),
each masking block (the pair of 1-D dilated convolution blocks 204s within the temporal convolution network separator 104 in fig. 2) comprising a first component (the first 204 with 1x1 conv 210, layer norm 208 in the pair of 1-D dilated convolution blocks 204s) and a second component (the second 204 in the pair of 1-D dilated convolution blocks 204s and outputted to an adder in fig. 2); and
transmitting information related the first masks produced by the series of masking blocks to a device (mask outputted from 104 to a multiplier 106 in figs. 1 and 2). except explicitly teaching wherein the first component generates a first mask for extracting speech and a second component that generates a second mask for extracting residual speech masked by the first mask, and each mask of the first mask and the second mask including mask values estimating an amount of speech present for each frame of the plurality of frames and each frequency band of the plurality of frequency bands.
Shin teaches an analogous field of endeavor by disclosing a computer-implemented method of mitigating audio artifacts (title and abstract, ln 1-8 and method steps in fig. 2 and solving an over-smoothing phenomenon that is a cause of speech distortion and degraded intelligibility of speech, para 5) and wherein Shin further teaches:
receiving, by the processor, audio data as a joint time-frequency representation over a plurality of frames and a plurality of frequency bands (the method is performed in a time-frequency domain, para 28 and represented as S(n, k) in equation 1, kth frequency and nth frame, i.e., time-frequency domain, para 39-41);
executing, by the processor, a digital model (training stage with DNN-based speech enhancement S123 and DNN-based residual error estimation S124 in fig. 2) for detecting speech from a feature vector of the audio data (outputting a restored speech S125 from extract feature S112 and S122 in fig. 2),
the digital model comprising a masking block (S113, S123, S114, S124 etc., in fig. 2),
the masking block comprising a first component (S113 in fig. 2) and a second component (S115 in fig. 2), and wherein the first component generates a first mask for extracting speech (outputted trained DNN to S123, para 74) and the second component that generates a second mask for extracting residual speech (outputted target-exaggerated residual error estimation to S124) masked by the first mask (the target-exaggerated residual error estimation applied on the output from S123), and
each mask of the first mask and the second mask including mask values (outputs as DNN trained value from DNN training S113 and as DNN trained target-exaggerated residual error estimation from residual error estimation DNN training S115, respectively) estimating an amount of speech present for each frame of the plurality of frames and each frequency band of the plurality of frequency bands (DNN-based speech enhancement
PNG
media_image1.png
22
13
media_image1.png
Greyscale
outputted from S123 based the DNN trained value and DNN-based residual error estimation
PNG
media_image2.png
19
12
media_image2.png
Greyscale
upon at least the signal obtained by the DNN-based speech enhancement, and outputted from DNN-based residual error estimation S124, respectively, para 74); and
transmitting information related the first masks produced by the series of masking blocks to a device (transmitting
PNG
media_image3.png
20
11
media_image3.png
Greyscale
as combination of
PNG
media_image1.png
22
13
media_image1.png
Greyscale
and
PNG
media_image2.png
19
12
media_image2.png
Greyscale
to the restore speech S125) for benefits of improving a performance of speech enhancement (by compensating the distortion due to over-smoothing, para 5, 36).
Therefore, it would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to have applied the masking block having the first component that generates the first mask for extracting speech and the second component that generates the second mask for extracting residual speech masked by the first mask, and each mask of the first mask and the second mask including mask values estimating the amount of speech present for each frame of the plurality of frames and each frequency band of the plurality of frequency bands, as taught by Shin, to each masking block comprising the first component and the second component in the computer-implemented method of mitigating audio artifacts, as taught by Kupryjanow, for the benefits discussed above.
Claim 14 has been analyzed and rejected according to claim 1 above and the combination of Kupryjanow and Shin further teaches a system for mitigating over-suppression of speech (Kupryjanow, and Shin, an apparatus in fig. 1, e.g., machine translation, a dialogue system, etc., para 4,), comprising: a memory; and one or more processors coupled to the memory and configured to perform the method of claim 1 (Kupryjanow, instructions stored in non-transitory machine-readable storage medium and executed by one or more processors, para 17, and Shin, computer-readable code or program stored in the CD, DVD, USB, etc., and implemented by a computer, para 97, and discussed in claim 1 above).
Claim 2: the combination of Kupryjanow and Shin further teaches, according to claim 1 above, the first component comprising a series of connected convolutional neural network CNN blocks with dilation (Kupryjanow, DNN training from the Y, S outputted extract feature S112 in fig. 2 and Shin, 1-D dilated convolution network 204).
Claim 3: the combination of Kupryjanow and Shin further teaches, according to claim 2 above, each CNN block of the series of connected CNN blocks comprising a CNN layer (Kupryjanow, convolution block 1D conv B with different dilation value), a batch normalization layer (layer norm 208), and an activation layer (layer 1x1 conv 210 that performs dimension reduction by controlling the depth of the input volume of the data, para 26, i.e., activation of the block 204).
Claim 4: the combination of Kupryjanow and Shin further teaches, according to claim 1 above, the first component comprising a CNN layer having a 1x1 filter (Kupryjanow, 1x1 conv 210 and Shin, DNN training S113).
Claim 8: the combination of Kupryjanow and Shin further teaches, according to claim 1 above, the first component of a first masking block of the series of masking blocks receiving the feature vector as an input, and the second component of the first masking block receiving an inverse of a result of applying the first mask to the feature vector (Kupryjanow, the 2nd 204 block performing inversed order of dilation level of the 1st 204 block in fig. 2 and Shin, the S124 and S115 further receiving signal Y from extract feature S122 and S112, respectively, and the signal Y is inverse of the DNN-based speech enhancement performed by S114 and S123, respectively).
Claim 9: the combination of Kupryjanow and Shin further teaches, according to claim 1 above, the digital model further comprising an input CNN block comprising a CNN layer (Kupryjanow, 210 as input to the block 204 in fig. 2) with other layer of lookahead stage (first convolution layer 304 after input layer in fig. 3, para 34-35, and normalization layer 208, para 26 in fig. 2) (Kupryjanow, layer norm 208 in fig. 2), except explicitly teaching with the other layer comprising with one lookahead filter, a batch normalization layer, and an activation layer.
An Official Notice is taken that an input stage of a neural network comprises CNN layer with one lookahead filter, a batch normalization layer, and an activation layer is notoriously well-known in the art for variety of benefits such as the batch normalization layer in the input for stabilizing and reducing internal covariate shift and allowing a higher learning rate, the lookahead filter added in the input stage for better controlling real-time latency and smoothing sequence prediction, and the activation layer for avoiding potential neural network collapses due to linear regression by introducing non-linearity into the neural network.
Therefore, it would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to have applied the lookahead filter, the batch normalization layer, and the activation layer at the input stage with the CNN layer, as taught by the well-known in the art, to the other layers with the CNN layer in the computer-implemente d method, as taught by the combination of Kupryjanow and Shin, for the benefits discussed above.
Claim 13: the combination of Kupryjanow and Shin further teaches, according to claim 1 above, the computer-implemented method further comprising:
training the digital model with a loss function (Kupryjanow, TCN trained with lost function calculated by using cost function or frequency SNR fSNR, para 38 and Shin, via the cost function mean-square error MSE) with non-linear penalty (Kupryjanow, the penalty is based on the cost function and by using a log operator in para 38-44, i.e., non-linear, and fSNR in para 38-44 and Shin, the MSE inherently gives non-linear cost function for penalty of inaccurate model to be trained, para 56) that penalizes speech either over suppression or under suppression (Kupryjanow, the mask is 0 if all noise, i.e., over-suppression, while mask is 1 if all speech, under-supression, para 28), except explicitly teaching that penalizing speech over suppression more than speech under suppression (Kupryjanow, the mask is 0 if all noise, i.e., over-suppression, while mask is 1 if all speech, under-supression, para 28).
It has been a recognized problem and need in the art, which may include a design need to solve the problem for enhancing speech from given noisy speech signal to be more close to the true speech, and there had been a finite number of identified, predictable potential solutions to achieve a relative enhancement of speech:
Penalizing the speech over-suppression more than speech under-suppression, for, e.g., relatively stronger noise level condition in the noisy speech signal,
Penalizing the speech under-suppression more than speech over-suppression, for, e.g., relatively weak noise level condition in the noisy speech signal,
it would have been obvious for one having ordinary skill in the art before the effective filing date of the claimed invention to have pursued the known potential solutions with a reasonable expectation of success or obvious to try, see MPEP 2141, III.
Therefore, it would have been obvious for one having ordinary skill in the art before the effective filing date of the claimed invention to have applied penalizing speech over-suppression more than speech under-suppression in the certain condition of noise level above in the obvious to try, to the non-linear penalization in the computer-implemented method, as taught by the combination of Kupryjanow and Shin, for the purpose above.
Claims 5, 15-16, 18, 20 are rejected under 35 U.S.C. 103 as being unpatentable over Kupryjanow (above) and in view of reference Shin (above) and Nyayate et al. (US 20210360349 A1, hereinafter Nyayate).
Claim 15 has been analyzed and partially rejected according to claims 1, 14 above and the combination Kupryjanow and Shin further teaches the rests of limitations of claim 15, including
the digital model comprising a masking block (Kupryjanow and Shin, digital model discussed in claim 1 above) comprising a first series of CNN blocks (Kupryjanow, dilated convolution blocks 204 and Shin, deep neural network in the DNN training S113) that generates a first mask (Kupryjanow, and Shin, output from the S113 and applied to S123) for extracting speech (Kupryjanow, the first 204 and output from the first 204 as input to the second 204 module in fig. 2, and Shin, generating extracted speech
PNG
media_image1.png
22
13
media_image1.png
Greyscale
and the discussed in claim 1 above) and a second block that generates a second mask for extracting residual speech masked by the first mask (Kupryjanow, the second 204 or later 204 and outputted to an adder in fig. 2, e.g., 204 block within the 204 stacks repeated n times, para 25, or the stack is repeated N times, e.g., eight stacked layers and three repeats, para 27, and Shin, residual error estimation DNN training S115 and discussed in claim 1 above), each CNN block of the first series of CNN blocks comprising a CNN layer (Kupryjanow, 1D conv B with different D values in fig. 2, and Shin, hidden layers, para 81), and each of the first mask and the second mask including mask values estimating an amount of speech present for each frame of the plurality of frames and each frequency band of the plurality of frequency bands (Kupryjanow and Shin, discussed in claim 1 above); and transmitting information related the first mask and the second mask (Kupryjanow, transmitting added value as mask from the output of different blocks 204s in fig. 2 and Shin, transmitting
PNG
media_image3.png
20
11
media_image3.png
Greyscale
as the combination of the
PNG
media_image1.png
22
13
media_image1.png
Greyscale
and the
PNG
media_image2.png
19
12
media_image2.png
Greyscale
as discussed in claim 1 above).
However, the combination of Kupryjanow and Shin does not explicitly teach wherein the second block that generates the second mask is GRU block having the GRU layer.
Nyayate teaches an analogous field of endeavor by disclosing a computer-readable non-transitory storage medium (title and abstract, ln 1-4 and including data storage 601, 605, 620 in fig. 6A) and wherein a digital model is disclosed (audio mask in fig. 2) and wherein the digital model comprising a masking block (Convolution2ds in 204) and the masking block comprising a first series of CNN blocks that generate a first mask (convolution2ds in 204 and outputted to concatenate layer 208) and a GRU block that generates a second mask (GRU in 206 and outputted to the concatenate layer 208) and wherein the GRU block having a GRU layer (GRU indicated by 206, para 57) for benefits of improving speech intelligence (by improving a quality of speech, para 50, and obtaining a high-accuracy speech, para 298 and for optimization, para 58).
Therefore, it would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to have applied the GRU block comprising the GRU layer, as taught by Nyayate, to the second block in the computer-readable, non-transitory storage medium, as taught by the combination of Kupryjanow and Shin, for the benefits discussed above.
Claim 5: the combination of Kupryjanow and Shin teaches, according to claim 1 above, the second component (discussed in claim 1 above), except explicitly teaching wherein the second component comprising a gated recurrent unit GRU block including a GRU layer.
Nyayate teaches an analogous field of endeavor by disclosing a computer-readable non-transitory storage medium (title and abstract, ln 1-4 and including data storage 601, 605, 620 in fig. 6A) and wherein a digital model is disclosed (audio mask in fig. 2) and further teaches, wherein the second component comprising a gated recurrent unit GRU block including a GRU layer (GRU 206 in fig. 2) for the benefits discussed in claim 15 above.
Therefore, it would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to have applied the GRU block comprising the GRU layer, as taught by Nyayate, to the second block in the computer-implemented method of mitigating audio artifacts, as taught by the combination of Kupryjanow and Shin, for the benefits discussed above.
Claim 16: the combination of Kupryjanow, Shin, and Nyayate further teaches, according to claim 15 above, wherein the masking block further comprising an additional block that derives a specific feature vector from a certain feature vector using the first mask and the second mask (Kupryjanow, after the first pair of 204 blocks in fig. 2, 204 block within the 204 stacks repeated n times, para 25, or the stack is repeated N times, e.g., eight stacked layers and three repeats, para 27, the subsequent pair of 204 block as claimed additional block, that outputs claimed specific feature vector based on the feature vector outputted from previous pair of 204 block, using the internal 204s within the current pair of 204 block in fig. 2, para 25, e.g., eight stacked layers and three repeats, para 27), the digital model comprising a series of masking blocks including the masking block (Kupryjanow, repeated pair of 204s of fig. 2 within the 204 stacks or repeated 204 stack N times, respectively, para 25, e.g., eight stacked layers and three repeats, para 27), a first masking block of the series of masking blocks (two 204s in fig. 2) receiving the feature vector (outputted from unit 210 in fig. 2), and each subsequent masking block of the series of masking blocks receiving the specific feature vector produced by a preceding masking block as an input (Kupryjanow, discussed above, e.g., subsequent pairs of 204 blocks within the 204 stack repeated n time, para 25 or the 204 stack repeated N times, e.g., eight stacked layers and three repeats, para 27).
Claim 18: the combination of Kupryjanow, Shin, and Nyayate further teaches, according to claim 15 above, the first series of CNN blocks having increasing dilation rates followed by decreasing dilation rates (the dilation is exponentially increased from bottom to top, D=1 to D= 16, followed by decreased dilation rate, D=16 to D=1 from top to down).
Claim 20 has been analyzed and rejected according to claims 15, 9 above.
Claims 12 is rejected under 35 U.S.C. 103 as being unpatentable over Kupryjanow (above) and in view of reference Shin (above) and Yu et al. (US 10283140 B1, hereinafter Yu).
Claim 12: the combination of Kupryjanow and Shin teaches, according to claim 1 above, the computer-implemented method further comprising:
receiving an input waveform in a time domain (Kupryjanow, capturing audio signal by microphones, para 14);
transforming the input waveform into raw audio data over a plurality of frequency bins and the plurality of frames (through STFT 102 to form Xk, n, k-frequency bin and n-frame in time-frequency domain, para 38-43 and Shin, obtaining audio data S(n, k) in equation 1, kth frequency and nth frame in time-frequency domain, para 39-41);
the joint time-frequency representation having an energy value for each time frame and each frequency band (Kupryjanow, Xk, n, k-frequency bin and n-frame in time-frequency domain and represented by magnitude spectrum and phase spectrum in fig. 2, and discussed above, and Shin, the signal S(n, k) and expressed in formula E(n, k) of para 40, with respect to a magnitude of the audio data, para 39-41, and the audio data or signal with the magnitude thus, having an energy value for each time frame and each frequency band is inherency and discussed in claim 1 above); and
generating the feature vector from the joint time-frequency representation (Kupryjanow, through encoder output including magnitude spectrum and phase spectrum in fig. 2, and Shin, via extract feature S122 and S112 in fig. 2 and discussed in claim 1 above), except explicitly teaching converting the raw audio data into the audio data by grouping the plurality of frequency bins into the plurality of frequency bands.
Yu teaches an analogous field of endeavor by disclosing a computer-implemented method (title and abstract, ln 1-11 and fig. 1A-1C) and wherein raw audio data are disclosed (reverberant audio 102 received by a microphone, col 4, ln 59-67, col 5, ln 1-5, in fig. 1A-1C) and further disclosed that the raw audio data are converted into the audio data (fully concatenated output by element 118 in figs. 1A-1C) by grouping the plurality of frequency bins into the plurality of frequency bands (based on clean speech has different time-frequency characteristics in different sub-bands, col 2, ln 26-32 and grouping frequency sub-bands into the full band in figs. 1A-1C or multiple frequency components produced from discrete Fourier transform DFT, col 5, ln 18-22 and grouped to individual sub-bands SUB-BAND 1, SUB-BAND2, …, SUB-BAND 4 in fig. 3A-3C, col 9, 1-20), the joint time-frequency representation (fully concatenated output outputted from element 110) having an energy value for each time frame and each frequency band (represented by time-frequency characteristics in different sub-bands, col 2, ln 26-32); and generating the feature vector from the joint time-frequency representation (through time-domain synthesizer to generate enhanced audio in figs. 1A-1C) for benefits of improving speech quality (specifically to degraded audio signal with a low SNR, col 1, ln 32-35, and by using sub-band DNN approach, figs. 1A-1C and by relatively high frequency resolution at low frequency sub-bands, col 7, ln 38-46).
Therefore, it would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to have applied converting the raw audio data into the audio data by grouping the plurality of frequency bins into the plurality of frequency bands, as taught by Yu, to transforming the input waveform into raw audio data over the plurality of frequency bins and the plurality of frames and the joint time-frequency representation in the computer-implemented method, as taught by the combination of Kupryjanow and Shin, for the benefits discussed above.
Examiner Comments/Allowable Subject Matters
Claims 6-7, 10-11, 17, 19 are rejected under 35 USC 112(b) due to the parent claims 1, 15 are rejected under 35 USC 112(b), as set forth above, for indefiniteness, which causes unclear in scope feature by feature and thus, in the claimed invention. Therefore, at this point, applying prior art rejection to the claims 6-7, 10-11, 17, 19 would not be considered proper, as it would have to be based on mere assumptions and considerable speculation about the scope of the claims, see MPEP 2173.06. It is noted that the rejection of claims 6-7, 10-11, 17, 19 under 35 USC 112(b) without application of prior art rejection is not an indication that the instantly amendment claims are patentable and it is also noted that, as best understood in view of the claim rejection under 35 USC 112(b), a prior art search has been updated by the examiner, and recorded in attached PTO-892 form.
Conclusion
Any inquiry concerning this communication or earlier communications from the examiner should be directed to LESHUI ZHANG whose telephone number is (571)270-5589. The examiner can normally be reached Monday-Friday 6:30amp-4:00pm 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, Vivian Chin can be reached at 571-272-7848. 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.
/LESHUI ZHANG/
Primary Examiner,
Art Unit 2695
