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 .
DETAILED ACTION
Claims 1-3 and 5-9 are pending. Claim 1 is independent.
Claims 2-3 and 5-9 depend from Claim 1.
Claim 4 is cancelled.
This Application was published as U.S. 2024/0194217.
Response to Amendment
Examiner thanks Applicant for the response filed on 28 Apr 2026 which has been correspondingly accepted and considered in this office action. Claims 1-3 and 5-9 are pending.
Response to Arguments
With regards to Claim Rejections - 35 USC § 103, Applicant has provided arguments, see pages 5-7, filed 28 Apr 2026 and amended claim 1. As a result, amendments to claims and arguments have been fully considered but they are not persuasive. Each argument will be addressed below in turn.
With regards to 35 USC § 103: Claim 1
Argument 1: Examiner’s interpretation of “accumulating an event number” is technically erroneous and exceeds the broadest reasonable interpretation (BRI).
Applicant's arguments filed 28 Apr 2026 page 6 are cited below:
The Examiner equates Yan's frame index (127/400) with the claimed ''event number." This interpretation is technically unsupported. In the field of data processing, there is a fundamental distinction between an index (ordinal data) and a count (cardinal data).
Yan' s '' 127/400'' is a position index. It identifies the address or sequence of a frame within a video. Summing frame indices (e.g., adding frame 127 to frame 128) yields no meaningful physical value and cannot be used to determine the content or status of a frame.
The claimed ''event number'' is a statistical total. It represents the accumulation of binary event indicators (Os and ls) within a specific frame. This number represents the density or frequency of acoustic events, which serves as the quantitative basis for setting a label.
By equating an ''index'' with an ''event count," the Examiner effectively collapses the distinction between ''where a frame is'' and ''what a frame contains." One of ordinary skill in the art would not understand the simple gathering of frame sequences in Yan to be equivalent to the statistical accumulation of internal event indicators as recited in claim 1.
In response, Examiner respectfully disagrees. The issue is accumulating an event number was properly interpreted under broadest reasonable interpretation (BRI) in light of the specification.
Under MPEP 2111, claims should be interpreted using broadest reasonable interpretation (BRI) in light of the specification. MPEP 2111 (II) states it is improper to import claim limitations from the specification.
Here, under BRI, a reasonable interpretation is that accumulating means to gather more over time, or to collect or bring together something. Accumulating can have both ordinal (index or order) and cardinal (count or total) interpretation. Such as be ordered as in the first, second, third, etc. of items gathered; or the total count or number of item accumulated. For an event, ordinary definition for an event is an occurrence, or something that happens at some time, and the broadest reasonable interpretation for an event number is a number of occurrences, or things that happens at some time. An event number also has ordinal and cardinal meanings, where the order of the events matter as well as the total number of events. This interpretation is consistent with Applicant specification which states, “accumulating the number of events in each frame” Par [0030] Accumulating the number of events under the broadest reasonable interpretation has a primarily cardinal interpretation that implies counting the number of events in each frame without worrying about the order. This is acceptable because Applicant can be its own lexicographer, and can broaden the claim interpretation beyond what is stated in the specification at the time of applicant’s filing. Thus, accumulating an event number has both ordinal and cardinal meanings under the broadest reasonable interpretation. Based on the ordinal definition, Applicant admits Yan teaches a “127/400” is a position index” (Applicant arguments page 6) which meets the ordinal definition under BRI of the claimed limitation because position or index in a sequence is an order and as explained in the 3 Feb 2026 office action, Mortensen in view of Yan, Kane, Steiner, and Sharma teaches all the claimed limitations of claim 1.
On the other hand, Applicant argues that “Summing frame indices (e.g., adding frame 127 to frame 128) yields no meaningful physical value and cannot be used to determine the content or status of a frame.” (Applicant arguments page 6) First of all, claim 1 does not mention sum. Summing and accumulating are not equivalent. Summing is a mathematical operation that has primarily cardinal value. This is because of the associative property of arithmetic where the order of a sum does not matter, i.e. 2+3=3+2. Furthermore, the claim does not provide any limitations on the “physical value”, “content”, or “status” of a frame. (id.) These limitations are not provided in the claim, and it is improper to import limitations from the specification that are not mentioned in the claim per MPEP 2111(II).
Applicant further argues that event number “represents the accumulation of binary event indicators (Os and ls) within a specific frame. This number represents the density or frequency of acoustic events, which serves as the quantitative basis for setting a label.” (Applicant arguments page 6) As mentioned above, the claim provides no details about how accumulate an event number for each frame other than to say it is done by the data capturing module. Thus, it is improper to import binary event indicators into the claim, when it is not mentioned in the claim. Furthermore, a video frame also includes audio, and thus under BRI, Yan teaches of frames is proper and meets the claimed limitations.
Argument 2: The relationship between event number, label, and training
Applicant's arguments filed 28 Apr 2026 page 6 are cited below:
As clarified by the current amendment, claim I discloses a specific data- driven process:
(1) Accumulation: An event number is derived by counting indicators.
(2) Determination: A label is determined based on that specific event number.
(3) Utilization: The database containing this label-to-count mapping is used to train a decision model.
The cited references fail to disclose this process. Yan's label or timestamp is an independent attribute stored in a lookup table for retrieval purposes. Yan does not teach or suggest that a label is generated as a result of an automated statistical accumulation of events within the frame.
In response, Examiner respectfully disagrees. The issue is how to interpret claims under the broadest reasonable interpretation of terms in light of the specification.
Under MPEP 2111, claims should be interpreted using broadest reasonable interpretation (BRI) in light of the specification. MPEP 2111 (II) states it is improper to import claim limitations from the specification.
Here, as previously discussed accumulation is more than counting because order matters, and under BRI, the order of the event number can be used to read on the claimed limitations.
As for determining, the claim does not provide details how to determine a label based on the event number. Thus, under BRI, any mechanism for determining a label based on an event number will read on the claim.
Argument 3: Failure to provide a motivation to combine with Sharma
Applicant's arguments filed 28 Apr 2026 pages 6-7 are cited below:
The Examiner relies on Sharma for metadata with timestamps. However, Sharma's metadata represents ''start and end times'' of speech segments, denoting a post- detection record. It does not disclose the ''event-indicative time-series'' (the Os and Is) disclosed by claim 1. Without the specific metadata structure of claim 1, there is no technical basis for the data capturing module to perform the claimed accumulation of event numbers.
Furthermore, combining Mortensen' s real-time audio stream with Yan' s retrieval-based indexing and Sharma' s data augmentation requires more than a simple gathering. It requires a fundamental redesign of the data format into the claimed event- based binary sequences. The Examiner's conclusion of obviousness relies on hindsight bias, using the Applicant's own disclosed architecture as a blueprint to patch together disparate functions from the cited art..
In response, Examiner respectfully disagrees. The issue is how to interpret claims about metadata under the broadest reasonable interpretation of terms in light of the specification.
Under MPEP 2111, claims should be interpreted using broadest reasonable interpretation (BRI) in light of the specification. MPEP 2111 (II) states it is improper to import claim limitations from the specification. MPEP 2143(I) provides examples of rationales to support a conclusion of obviousness such as:
(A) Combining prior art elements according to known methods to yield predictable results;
(B) Simple substitution of one known element for another to obtain predictable results;
(C) Use of known technique to improve similar devices (methods, or products) in the same way;
(D) Applying a known technique to a known device (method, or product) ready for improvement to yield predictable results;
(E) "Obvious to try" – choosing from a finite number of identified, predictable solutions, with a reasonable expectation of success;
(F) Known work in one field of endeavor may prompt variations of it for use in either the same field or a different one based on design incentives or other market forces if the variations are predictable to one of ordinary skill in the art;
(G) Some teaching, suggestion, or motivation in the prior art that would have led one of ordinary skill to modify the prior art reference or to combine prior art reference teachings to arrive at the claimed invention
Here, the claim has no specific definition for metadata. As was stated in the final office of 4 Sep 2025, “applicant has not provided a special definition of metadata, so metadata has its plain meaning of data that defines and characterizes other data. Thus, Yan teaches metadata from videos is also metadata about audio signals.” (4 Sep 2025 Final office action Page 4) No arguments were received with respect to this argument, and no specific definition of metadata has been provided. Thus, the office maintains that definition of metadata as previously discussed. Furthermore, while Applicant argues lack of motivation, Applicant does not provide any arguments that would refute the 35 USC § 103 conclusion of obviousness. For completeness, it is maintained that 3 Feb 2026 office action provided some teaching, suggestion, or motivation that would lead a person of ordinary skill in the art to combine Mortensen in view of Yan, Kane, and Steiner with Sharma.
On the other hand, Applicant argues that Sharm “does not disclose the ''event-indicative time-series'' (the Os and Is) disclosed by claim 1. Without the specific metadata structure of claim 1, there is no technical basis for the data capturing module to perform the claimed accumulation of event numbers.” (Applicant arguments page 6-7) First, claim does not mention any “event-indicative time-series (the Os and Is)” or any specific structure of metadata. Under BRI, any metadata structure is acceptable if it reads on the claims. As discussed above, the metadata structure was described above and in the 3 Feb 2026 office action, so the claimed limitations have been met.
Applicant further argues that “combining Mortensen' s real-time audio stream with Yan' s retrieval-based indexing and Sharma' s data augmentation requires more than a simple gathering. It requires a fundamental redesign of the data format into the claimed event-based binary sequences.” (Applicant arguments page 7) It must first be noted that claim 1 does not mention binary, any sequence, or data format. These limitations are not in the claim, and it is improper to argue limitations that are not in the claim that may be in the specification. So these arguments are not relevant.
Finally, Applicant argues that “Examiner's conclusion of obviousness relies on hindsight bias, using the Applicant's own disclosed architecture as a blueprint to patch together disparate functions from the cited art.” (Applicant arguments page 7) Office action of 3 Feb 20206 provides motivation by stating “Sharma teaches “data augmentation may allow for the generation of new training data for a machine learning system by augmenting existing data to represent new conditions” (Par [0002]) which increases the accuracy of the voice activity detection invention of Mortensen in view of Yan, Kane, and Steiner” (3 Feb 2026 office action page 10) This is not hindsight, because it combines the new training data as taught by Mortensen in view of Yan, Kane, and Steiner to be combined with the augmentation system as taught by Sharma. References that teach training data, and combining them with a system that allows for new training data is not hindsight.
Thus, for all the reasons stated above, applicant arguments are not persuasive.
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.
Claims 1, and 6-9 are rejected under 35 U.S.C. 103 as being unpatentable over Mortensen et al. (US2019/0355383 hereinafter Mortensen) in view of Yan (US2023/0086735 hereinafter Yan), Kane (US2022/0201121 hereinafter Kane), Steiner et al. (US2023/0062377 hereinafter Steiner) in further view of Sharma et al.(US2023/0230599 hereinafter Sharma).
With regards to claim 1, Mortensen teaches:
A data processing method for acoustic event comprising performing a plurality of steps by a processor, [Fig 3, item 304] wherein the plurality of steps comprises: establishing a simulated acoustic frequency event module, [Fig 5, item 500-501] a data capturing module, [Fig 5, item 500-501] and a sound application decision module in a software manner, [Fig 5, item 516, Par [0229]]
wherein the simulated acoustic frequency event module comprises: a plurality of frequency band filter modules, [Fig 5, item 501, 502, and 504] a plurality of energy estimation modules connecting to the plurality of frequency band filter modules, [Fig 5, item 500, 501, and 504] and a plurality of frequency event quantizers connecting to the plurality of energy estimation modules [Fig 11, item 1104 where Fig 11 shows a one channel configuration where an analog digital converter (ADC) is a frequency event quantizer; however Mortensen teaches “More channels or pairs of channels can be used to detect different types of voices to improve detection and/or to detect voices present in different audio streams”(Par [0004])]
setting at least one of the plurality of frequency band filter modules, the plurality of energy estimation modules and the plurality of frequency event quantizers according to a simulated hardware parameter; [Mortensen Par [0098] teaches frequency band filter module can have simulated hardware parameters such that “Any suitable filter can be used for reducing the bandwidth of the incoming audio stream to just the first frequency band, e.g., the frequency band of interest which covers a reasonable number of vowel F1 formant frequencies”]
inputting a sound signal [Fig 5 incoming audio stream and Fig 11 item 1102] to the plurality of frequency band filter modules [Fig 5 and 11] and obtaining a plurality of metadata from the plurality of frequency event quantizers, wherein the sound signal is an analog electric signal and the plurality of metadata is digital signals; [Mortensen teaches obtaining a plurality of metadata or digital from the plurality of ADCs or frequency event quantizers in Fig 11 with output into decision module (516), where more than one channel can be used (Par [0004]) for the plurality of frequency event quantizers. Analog electrical signals can be obtained from the plurality of frequency band filters (502) shown in Fig 5 for 2 channels with output into decision module(516)]
wherein the simulated hardware parameter is configured to be assigned to the plurality of frequency event quantizers, and the simulated hardware parameter comprises a data dynamic range, [Mortensen Fig 20 item 612, Par [0177] teaches “crest detector has a top tracker which tracks the peaks of the signal and a bottom tracker which tracks the quiet periods of the signal. The difference between these two is the modulation index of the signal” (Par [0112]) where the modulation index is the dynamic data range]
a number of channels, [Mortensen Fig 5 and 11] a number of the plurality of frequency event quantizers is equal to the number of channels, [Mortensen Fig 11]
With regards to claim 1, Mortensen fails to teach:
dividing each of the plurality of metadata into a plurality of frames according to a time interval by the data capturing module, wherein each of the plurality of frames has a timestamp;
accumulating an event number of each of the plurality of frames by the data capturing module, setting a label of each of the plurality of frames according to the event number, and storing the plurality of frames, the event number and the label in a database; and
training a decision model by the sound application decision module according to the database and a sound application;
wherein the label is determined based on the event number and
With regards to claim 1, Yan teaches:
dividing each of the plurality of metadata into a plurality of frames according to a time interval by the data capturing module, [Yan Fig 4a teaches visual relationship system (102) or data capturing module “dividing the video 114 into subsections, e.g., 30 second clips of video, and then selecting a representative frame of each subsection to be the key frame 11” (Par [0057]) where video consists of a plurality of digital signals or metadata which is divided into a plurality of frames according to a time interval by the data capturing module]
wherein each of the plurality of frames has a timestamp; [Fig 4a item 207, Par [0057]]
accumulating an event number of each of the plurality of frames by the data capturing module, [Yan Par [0057] teaches “a key frame 115 can be associated with a timestamp 207 that indicates a frame number of a total number of frames of the video 114, for example, 127/400, where the key frame 115 appears at the 127th frame of a total of 400 frames of the video 11”]
setting a label of each of the plurality of frames according to the event number, and [Yan Par [0057] teaches a “label assigned to the key frame 115” … [and] “a key frame 115 can be associated with a timestamp 207 that indicates a frame number of a total number of frames of the video 114”]
storing the plurality of frames, the event number and the label in a database; and [Yan Par [0057] teaches “timestamp 207 can be associated with the key frame 115 in an index/table,” “timestamp 207 can be a label assigned to the key frame 115” and “key frame 115 can be associated with a timestamp 207 that indicates a frame number of a total number of frame,” where index/table can be to develop “scene graph index can be a lookup table that identifies each key frame and its corresponding scene graph and timestamp, as depicted in FIG. 2A” which is stored in scene graph database (118).
wherein the label is determined based on the event number and [Yan Par [0057] teaches a “label assigned to the key frame 115” … [and] “a key frame 115 can be associated with a timestamp 207 that indicates a frame number of a total number of frames of the video 114” where the label is determined based on the event number]
It would be obvious to one of ordinary skill in the art to combine the digital signals from the voice activity detection system and as taught by Mortensen with the creating of frames and labels for the digital signals as taught by Yan. The motivation to combine the teachings of a Mortensen with the teachings of Yan is because Yan teaches creating an “index/table” (Par [0057]) with a timestamp and label that makes it easier to search” (Par [0015]) which increases the capabilities of the invention of Mortensen]
With regards to claim 1, Mortensen in view of Yan fails to teach:
training a decision model by the sound application decision module according to the database and a sound application;
wherein the label is stored in the database used for training the decision model;
With regards to claim 1, Kane teaches:
training a decision model by the sound application decision module according to the database and a sound application; wherein the label is stored in the database used for training the decision model; [Kane Par [0048] teaches training “various machine learning models” which is a decision model by computing features such as “acoustic measurements, such as pitch, energy, voice activity detection, speaking rate, turn-taking characteristics, and time-frequency spectral coefficient” which are sound applications, and using “labeled training data in the behavior training database 116.”
It would be obvious to one of ordinary skill in the art to combine the digital signals from the voice activity detection system and as taught by Mortensen with the creating of frames and labels for the digital signals as taught by Yan, with the training a machine learning model and storing in a particular database as taught by Kane. The motivation to combine the teachings of a Mortensen and Yan with the teachings of Kane is because Kane teaches “compute features used as input to machine learning models (such models may be developed offline and, once developed, can make inferences in real-time” (Par [0034]) which increases the capabilities of the invention of Mortensen in view of Yan to a real time process]
With regards to claim 1, Mortensen in view of Yan and Kane fails to teach:
a bit width, a time resolution, a time interval and the plurality of frequency event quantizers are configured to output a first value representing that an event occurs when an energy of an input signal is greater than a threshold, and output a second value representing that the event does not occur when the energy of the input signal is smaller than the threshold.
With regards to claim 1, Steiner teaches:
a bit width, [Steiner Par [0028] teaches “bit width of the ADC can be, for example, 8 bit or higher]
a time resolution, [Steiner Par [0028] teaches “ADC sampling frequency may be set to 5 msps” where time resolution may be calculated as the inverse of the sampling frequency which equals 20ns]
a time interval and [Steiner teaches “predetermined observation window has a start time and an end time” (Par [0028] … “For example, 10 us or 20 us are possible observation window durations” (Par [0031]
the plurality of frequency event quantizers are configured to output a first value representing that an event occurs when an energy of an input signal is greater than a threshold, [Steiner teaches “If the Euclidian distance dNoTouch is equal to or greater than the first threshold value, the evaluation processing circuit 207 may detect a touch event” (Par [0049]) where the value is a touch event]
and output a second value representing that the event does not occur when the energy of the input signal is smaller than the threshold. [Steiner teaches “If the Euclidian distance dNoTouch is less than the first threshold value Threshold 1, the evaluation processing circuit 207 may detect a no-touch event” (Par [0049]) where the value is a no touch event]
It would be obvious to one of ordinary skill in the art to combine voice activity detection system and as taught by Mortensen in view of Yan and Kane with the analog to digital converter (ADC) and touch sensor as taught by Steiner. The motivation to combine the teachings of a Mortensen, Yan, and Kane with the teachings of Steiner is because Steiner teaches using a touch sensor that “relies on the transmission of an ultra-sonic signal and the reception and processing of the reflected waveform from the touch surface” (Steiner Par [0001]) which increases the capabilities of the invention of Mortensen in view of Yan and Kane in an expanded acoustic range which allows the potential for other applications such as a touch sensor]
With regards to claim 1, Mortensen in view of Yan, Kane, and Steiner fails to teach:
wherein each of the plurality of metadata comprises an event-indicative time-series including a plurality of event indicators, each event indicator indicating occurrence or non-occurrence of an event at a corresponding time;
With regards to claim 1, Sharma teaches:
wherein each of the plurality of metadata comprises an event-indicative time-series including a plurality of event indicators, each event indicator indicating occurrence or non-occurrence of an event at a corresponding time; [Sharma Fig 4 teaches augmentation process (10) creates event indicators where speech is active or inactive (Par [0067]) which are indicative of occurrence or non-occurrence of an event where “augmentation process 10 may generate acoustic metadata with timestamps indicating portions of first device speech signal 402 that include speech activity (e.g., start and end times for each portion)” (Par [0067])]
It would be obvious to one of ordinary skill in the art to combine voice activity detection system and as taught by Mortensen in view of Yan, Kane, and Steiner with the data augmentation system as taught by Sharma. The motivation to combine the teachings of a Mortensen, Yan, Kane, and Steiner with Sharma is because Sharma teaches “data augmentation may allow for the generation of new training data for a machine learning system by augmenting existing data to represent new conditions” (Par [0002]) which increases the accuracy of the voice activity detection invention of Mortensen in view of Yan, Kane, and Steiner]
With regard to claim 6, Mortensen in view of Yan, Kane, Steiner, and Sharma teaches:
All the limitations of claim 1
further comprising adjusting the simulated hardware parameter by the processor according to an accuracy of the decision model, [Mortensen Par [0222] teaches “output indicating whether voice activity detection can have discrete levels indicating varying probabilities that voice activity is present” where outputting probabilities can determine the accuracy of the decision model]
an adjusted record of the simulated hardware parameter. [Mortensen Fig 25 Par [0197] teaches a registry file map that “serves as storage for audio input, parameters for the computations, calculated values as well as mapped locations for CSR's (control status registers), etc.” which records parameters and can be adjusted]
With regards to claim 6, Mortensen in view of Yan fails to teach:
an accuracy threshold, and
With regards to claim 6, Kane teaches:
an accuracy threshold, and [Kane Par [0079] teaches “machine learning model outputs is typically a probability, so this needs to be binarized by applying a threshold.”
It would be obvious to one of ordinary skill in the art to combine the teachings of Mortensen in view of Yan that teaches determining the accuracy of the decision model with probability with the teachings of Kane that teaches using a threshold on the probability outputs from the machine learning model. The motivation to combine the teachings of Mortensen and Yan with the teachings of Kane is because Kane teaches “compute features used as input to machine learning models (such models may be developed offline and, once developed, can make inferences in real-time” (Par [0034]) which increases the capabilities of the invention of Mortensen in view of Yan to a real time process]
With regard to claim 7, Mortensen in view of Yan, Kane, Steiner, and Sharma teaches:
All the limitations of claim 1
wherein the sound application comprises a voice activity detection, [Mortensen Fig 5, Par [0097]]
a keyword spotting, [Mortensen Par [0204] teaches “frequency bands can be adjusted based on one or more pre-defined utterance/phrase. Specifically, the frequency band of a particular channel can be tuned for one or more specific vowels of interest. For instance, various voice activated programs triggers when a user utters or say a particular keyword” which describes keyword spotting]
an acoustic environment identification, [Mortensen Par [0146] teaches “processor performing the process triggered by the output signal(s) of the decision modules can in some cases select a suitable process based on the information inferred. The resulting system can be more aware of the environment near these audio capturing devices, and thus provide contextually aware processes in response to the outputs of the voice activity detector” which describes an acoustic environment identification]
an acoustic abnormal sound detection, and [Mortensen Par [0196] teaches “updating of one or more parameters based on environmental conditions (e.g., level of noise in the environment)” where the level of noise can abnormal sound]
an output number of the decision model is associated with the sound application. [Mortensen teaches “the decision module 516 can be provide a counting filter (absorbing the low pass filtering module 514)” (Par [0104]) where the “decision module 516 is configured to output a signal to indicate that voice activity is detected in the audio stream” (Par [0104]) and the count is the number of detected voice activities]
With regards to claim 7, Mortensen in view of Yan fails to teach:
the decision model is a fully connected neural network,
With regards to claim 7, Kane teaches:
the decision model is a fully connected neural network, [Kane Par [0078] teaches training the decision model by the sound application decision module according to the database and the sound application via “Supervised machine learning using neural networks.”
It would be obvious to one of ordinary skill in the art to combine the digital signals from the voice activity detection system and as taught by Mortensen with the creating of frames and labels for the digital signals as taught by Yan, with the training a machine learning model using supervised learning as taught by Kane. The motivation to combine the teachings of a Mortensen and Yan with the teachings of Kane is because Kane teaches “compute features used as input to machine learning models (such models may be developed offline and, once developed, can make inferences in real-time” (Par [0034]) which increases the capabilities of the invention of Mortensen in view of Yan to a real time process]
With regards to claim 7, Mortensen in view of Yan and Kane fails to teach:
an ultrasonic vibration detection, and
With regards to claim 7, Steiner teaches:
an ultrasonic vibration detection, and [Steiner Fig 1 item 106-7, Par [0021] teaches “receiver (RX) 106 configured to receive reflected ultra-sound signals, and a sensor circuit 107 (e.g., an application specific integrated circuit (ASIC)) configured to generate the ultra-sound signals for transmission by the transmitter” which describes ultrasonic vibration detection.
It would be obvious to one of ordinary skill in the art to combine voice activity detection system and as taught by Mortensen in view of Yan and Kane with the analog to digital converter (ADC) and touch sensor as taught by Steiner. The motivation to combine the teachings of a Mortensen, Yan, and Kane with the teachings of Steiner is because Steiner teaches using a touch sensor that “relies on the transmission of an ultra-sonic signal and the reception and processing of the reflected waveform from the touch surface” (Steiner Par [0001]) which increases the capabilities of the invention of Mortensen in view of Yan and Kane in an expanded acoustic range which allows the potential for other applications such as a touch sensor]
With regards to claim 8, Mortensen in view of Yan, Kane, Steiner, and Sharma teaches:
All the limitations of claim 1
With regards to claim 8, Mortensen in view of Yan fails to teach:
wherein training the decision model by the sound application decision module according to the database and the sound application is a supervised learning.
With regards to claim 8, Kane teaches:
wherein training the decision model by the sound application decision module according to the database and the sound application is a supervised learning. [Kane Par [0078] teaches training the decision model by the sound application decision module according to the database and the sound application via “Supervised machine learning using neural networks.”
It would be obvious to one of ordinary skill in the art to combine the digital signals from the voice activity detection system and as taught by Mortensen with the creating of frames and labels for the digital signals as taught by Yan, with the training a machine learning model using supervised learning as taught by Kane. The motivation to combine the teachings of a Mortensen and Yan with the teachings of Kane is because Kane teaches “compute features used as input to machine learning models (such models may be developed offline and, once developed, can make inferences in real-time” (Par [0034]) which increases the capabilities of the invention of Mortensen in view of Yan to a real time process]
With regard to claim 9, Mortensen in view of Yan, Kane, Steiner, and Sharma teach:
All the limitations of claim 1
wherein a value setting of the simulated hardware parameter is associated with the sound application. [Mortensen teaches “More channels or pairs of channels can be used to detect different types of voices to improve detection and/or to detect voices present in different audio streams”(Par [0004]) where the number of channels is a value setting of the simulated hardware parameter is associated with the sound application]
Claim 2 is rejected under 35 U.S.C. 103 as being unpatentable over Mortensen et al. (US2019/0355383) in view of Yan (US2023/0086735), Kane (US2022/0201121), Steiner et al. (US2023/0062377), and Sharma et al.(US2023/0230599) in further view of Hussin et al. (.S. F. Hussin, G. Birasamy, and Z. Hamid, “Design of Butterworth Band-Pass Filter”, Politeknik & Kolej Komuniti Journal of Engineering and Technology, vol. 1, no. 1, pp. 32–46, Nov. 2016) hereinafter Hussin.
With regards to claim 2, Mortensen in view of Yan, Kane, Steiner, and Sharma teaches:
All the limitations of claim 1
wherein the simulated hardware parameter is configured to be assigned to the plurality of frequency band filter modules, and the simulated hardware parameter comprises, a frequency lower limit, [Mortensen Fig 11, item 1108 bandpass filter] a frequency upper limit, [Mortensen Fig 11, item 1108 bandpass filter] a filter bandwidth, [Mortensen Fig 11, item 1108 bandpass filter] a filter method, [Mortensen Fig 11, item 1108 bandpass filter] and a number of channels, [Mortensen Fig 5 and 11] and a number of the plurality of frequency band filter modules is equal to the number of channels [Mortensen Fig 5 and 11]]
With regards to claim 2, Mortensen in view of Yan, Kane, Steiner, and Sharma fails to teach:
a filter gain
a central frequency,
a filter order,
With regards to claim 2, Hussin teaches:
a filter gain [Hussin Fig 4, pg 34-35 teaches design “Bandpass Butterworth filter need to design in this paper must have the characteristic as shown in figure 4” which describes gain and frequency parameters]
a central frequency, [Hussin pg 33 teaches “Transmitted and received signals have to be filtered at a certain center frequency with a specific bandwidth” where central or center frequency will depend on lower and upper frequency bounds or bandwidth]
a filter order, [Hussin pg 41 teaches “Procedure of designing LPF and HPF is divided into two parts, the first part is finding the required order of the filter” where the low and high pass filter (LPF and HPF respectfully) are determined based on the order of the filter.
It would be obvious to one of ordinary skill in the art to combine voice activity detection system and as taught by Mortensen in view of Yan, Kane, Steiner, and Sharma with the bandpass filter as taught by Hussin. The motivation to combine the teachings of a Mortensen, Yan, Kane, Steiner, and Sharma with the teachings of Hussin is because “By providing suitable bandpass filters and channels, formant filtering can be used to detect bird sounds (or bird speech)” (Mortensen Par [0230]) which increases the capabilities of the invention of Mortensen in view of Yan, Kane, Steiner, and Sharma to better detect sounds]
Claims 3 and 5 are rejected under 35 U.S.C. 103 as being unpatentable over Mortensen et al. (US2019/0355383) in view of Yan (US2023/0086735), Kane (US2022/0201121), Steiner et al. (US2023/0062377), and Sharma et al.(US2023/0230599) in further view in further view of Mansour et al. (US2019/0028130 hereinafter Mansour).
With regards to claim 3, Mortensen in view of Yan, Kane, Steiner, and Sharma teaches:
All the limitations of claim 1
wherein the simulated hardware parameter is configured to be assigned to the plurality of energy estimation modules, and the simulated hardware parameter comprises an energy gain, [Mortensen Fig 20 item 606, Fig 23, Par [0183] where the running average is an energy gain] an energy threshold, [Mortensen Fig 20, item 614, Par [01777]] and a number of channels, [Mortensen Fig 5 and 11] a number of the plurality of energy estimation modules is equal to the number of channels, [Mortensen Fig 5 and 11]
With regards to claim 3, Mortensen in view of Yan, Kane, Steiner, and Sharma fails to teach:
and the plurality of energy estimation modules are implemented by a waveform rectifier.
With regards to claim 3, Mansour teaches:
and the plurality of energy estimation modules are implemented by a waveform rectifier. [Mansour Fig 1, item 140, Par [0040] teaches calculation unit (140) is an energy estimation module for signals that can “represent energy, a power and in general, the presence of oscillations or amplitude in the particular associated band-pass filtered signal” that is implemented by rectifier (R1 through Rn) for waveforms from the band pass filter.
It would be obvious to one of ordinary skill in the art to combine voice activity detection system and as taught by Mortensen in view of Yan, Kane, Steiner, and Sharma with the rectifier as taught by Mansour. The motivation to combine the teachings of a Mortensen, Yan, Kane, Steiner, and Sharma with the teachings of Mansour is because “signal parameters can be calculated in the calculation step which each represent an energy and/or power of a band-pass filtered signal. Such an embodiment of the proposed approach offers the advantage of providing highly meaningful information by means of the signal parameter, which enables an easily implemented inference as to the relevance or the information contained in the band-pass filtered signa” (Mansour Par [0016]) which increases the capabilities of the invention of Mortensen in view of Yan, Kane, Steiner, and Sharma to provide information in the band pass signal processing]
With regards to claim 5, Mortensen in view of Yan, Kane, Steiner, and Sharma teaches:
All the limitations of claim 1
before inputting the sound signal to the plurality of band filter modules, further comprising: establishing an amplifier in the software manner; and [Mortensen Fig 11 item 1104, Par [0177] teaches “it is possible to not only implement the model in software embodied in non-transient computer-readable medium, it is possible to implement the model in hardware”]
inputting an audio stream into the amplifier to generate the sound signal, [Mortensen Fig 11 item 1104, Par [0150]]
With regards to claim 5, Mortensen in view of Yan, Kane, Steiner, and Sharma fails to teach:
wherein an output of the plurality of frequency band filter modules and an output of the plurality of energy estimation modules are one of voltage, current, and charge.
With regards to claim 5, Mansour teaches:
wherein an output of the plurality of frequency band filter modules and an output of the plurality of energy estimation modules are one of voltage, current, and charge. [Mansour Par [0040] teaches each “signal parameters can be calculated in the calculation step which each represent an energy and/or power of a band-pass filtered signal” where voltage, current, and charge can be calculated given power and energy, since power can be calculated from Ohm’s law as the product of current and voltage, and current is equal to the change in charge over time.
It would be obvious to one of ordinary skill in the art to combine voice activity detection system and as taught by Mortensen in view of Yan, Kane, Steiner, and Sharma with the rectifier as taught by Mansour. The motivation to combine the teachings of a Mortensen, Yan, Kane, Steiner, and Sharma with the teachings of Mansour is because “signal parameters can be calculated in the calculation step which each represent an energy and/or power of a band-pass filtered signal. Such an embodiment of the proposed approach offers the advantage of providing highly meaningful information by means of the signal parameter, which enables an easily implemented inference as to the relevance or the information contained in the band-pass filtered signal” (Mansour Par [0016]) which increases the capabilities of the invention of Mortensen in view of Yan, Kane, Steiner, and Sharma to provide information in the band pass signal processing]
Conclusion
Applicant's amendment necessitated the new ground(s) of rejection presented in this Office action. Accordingly, THIS ACTION IS MADE FINAL. See MPEP § 706.07(a). Applicant is reminded of the extension of time policy as set forth in 37 CFR 1.136(a).
A shortened statutory period for reply to this final action is set to expire THREE MONTHS from the mailing date of this action. In the event a first reply is filed within TWO MONTHS of the mailing date of this final action and the advisory action is not mailed until after the end of the THREE-MONTH shortened statutory period, then the shortened statutory period will expire on the date the advisory action is mailed, and any nonprovisional extension fee (37 CFR 1.17(a)) pursuant to 37 CFR 1.136(a) will be calculated from the mailing date of the advisory action. In no event, however, will the statutory period for reply expire later than SIX MONTHS from the mailing date of this final action.
Any inquiry concerning this communication or earlier communications from the examiner should be directed to Joseph J Yamamoto whose telephone number is (571)272-4020. The examiner can normally be reached M-F 1000-1800 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, Bhavesh Mehta can be reached at 571-272-7453. 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.
JOSEPH J. YAMAMOTO
Examiner
Art Unit 2656
/BHAVESH M MEHTA/Supervisory Patent Examiner, Art Unit 2656
Prosecution Insights