DETAILED ACTION
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 .
This Office action is in response to the application filed on 10/01/2024.
Claims 1-20 are presented for examination.
Information Disclosure Statement
The information disclosure statement (IDS) submitted 09/22/2026 is in compliance with the provisions of 37 CFR 1.97. Accordingly, the information disclosure statement is considered by the examiner.
Claim Rejections - 35 USC § 112
The following is a quotation of the first paragraph of 35 U.S.C. 112(a):
(a) IN GENERAL.—The specification shall contain a written description of the invention, and of the manner and process of making and using it, in such full, clear, concise, and exact terms as to enable any person skilled in the art to which it pertains, or with which it is most nearly connected, to make and use the same, and shall set forth the best mode contemplated by the inventor or joint inventor of carrying out the invention.
The following is a quotation of the first paragraph of pre-AIA 35 U.S.C. 112:
The specification shall contain a written description of the invention, and of the manner and process of making and using it, in such full, clear, concise, and exact terms as to enable any person skilled in the art to which it pertains, or with which it is most nearly connected, to make and use the same, and shall set forth the best mode contemplated by the inventor of carrying out his invention.
Claims 1-20 are rejected under 35 U.S.C. 112(a) or 35 U.S.C. 112 (pre-AIA ), first paragraph, as failing to comply with the written description requirement. The claim(s) contains subject matter which was not described in the specification in such a way as to reasonably convey to one skilled in the relevant art that the inventor or a joint inventor, or for applications subject to pre-AIA 35 U.S.C. 112, the inventor(s), at the time the application was filed, had possession of the claimed invention.
Claims 1, 2, and 14-16 recite “comparing the second plurality of outputs generated by the first machine learning model to the second output generated by the second machine learning model to determine whether to use (i) an output of the second plurality of outputs generated by the first machine learning model or (ii) the second output generated by the second machine learning model to generate a dynamic conversational response” in claim 1; “comparing the plurality of outputs to the second output to determine whether to use an output of the plurality of outputs or the second output to generate a dynamic conversational response” in claim 2; “comparing the generated aggregated measure of performances to identify a model with the greatest aggregate measure of performance” in claims 14-15; and “comparing the set of outputs to the second output to determine whether to use (i) at least one of the set of outputs or (ii) the second output to generate a dynamic conversational response” in claim 16. However, the Specification does not describe or provide written description support for this limitation.
Claim Rejections - 35 USC § 103
In the event the determination of the status of the application as subject to AIA 35 U.S.C. 102 and 103 (or as subject to pre-AIA 35 U.S.C. 102 and 103) is incorrect, any correction of the statutory basis (i.e., changing from AIA to pre-AIA ) for the rejection will not be considered a new ground of rejection if the prior art relied upon, and the rationale supporting the rejection, would be the same under either status.
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 2-3, 6-8 and 10-20 are rejected under 35 U.S.C. 103 as being unpatentable over Aguilar Alas et al. (US 2021/0104245 A1), in view of Zhou (US 2019/0206397 A1), Harding et al. (US 11,038,821 B1).
As to claim 2, Aguilar Alas discloses the invention as claimed, including a method for
a dynamic conversational response using ensemble prediction based on a plurality of machine learning models (Fig. 6; ¶0022--¶0023; ¶0086; ¶0098), the method comprising:
receiving a first user action during a conversational interaction via a user interface
(¶0032, “The system(s) 120 receives (134) text data representing the utterance. The system(s) 120 determines (136) at least one lexical feature vector using the text data”; ¶0037, “system(s) 120 receives (150) input audio data representing an utterance”; Claim 4, “wherein receiving the input audio data comprises receiving the second utterance spoken by a first user”);
determining, based on the first user action, a first feature input for a first model (Fig. 6, 615), wherein the first model is an ensemble of a plurality of separately trained models (¶0033, “the first ML model is trained to process acoustic features derived from the training audio data, while the second ML model is trained to process lexical features derived from text data corresponding to the training audio data”; ¶0039, “the system(s) 120 may determine a first portion of the input audio data corresponding to a first word of the utterance and a second portion of the input audio data corresponding to a second word of the utterance”; ¶0115, “The system trains two separate machine learning (ML) models (615 and 630) focused on different modalities. For example, ML model 615 is trained to process acoustic features derived from the training audio data 602, while ML model 630 is trained to process lexical features derived from text data 606”);
inputting the first feature input into the first model (Fig. 6, 615) to determine a plurality of outputs from the first model (¶0038, “processes (154) the feature vector using the trained model (determined during the training operations of FIG. 1A) to determine model output data”; ¶0039, “The system(s) 120 determines a first feature vector using the first portion of input audio data and a second feature vector using the second portion of the input audio data. The system(s) 120 processes the first and second feature vectors using the trained model to determine first and second model output data”);
determining, based on the plurality of outputs, a second feature input for a second model (Fig. 6, 630) (¶0038; ¶0039, “the system(s) 120 may determine a first portion of the input audio data corresponding to a first word of the utterance and a second portion of the input audio data corresponding to a second word of the utterance. The system(s) 120 determines a first feature vector using the first portion of input audio data and a second feature vector using the second portion of the input audio data”; -¶0044, “determine time data indicating when the portion of the input audio data is received by the device 110. The output data may include the text data corresponding to the portion of input audio data, the time data”);
inputting the second feature input into the second model (Fig. 6, 620) to determine a second output (¶0032, “The system(s) 120 receives (134) text data representing the utterance. The system(s) 120 determines (136) at least one lexical feature vector using the text data. The ASR component 250, as described below, may generate an ASR output including text data representative of an utterance represented in the training audio data”; ¶0038, “The trained model is trained using acoustic and lexical features to detect sentiment in audio data. The model output data may indicate a score or probability of utterance corresponding to one of four sentiment categories (e.g., anger, happiness, sadness, and neutral). The system(s) 120 determines (156), using the model output data, a sentiment category for the utterance”; ¶0039, “The system(s) 120 determines a first feature vector using the first portion of input audio data and a second feature vector using the second portion of the input audio data. The system(s) 120 processes the first and second feature vectors using the trained model to determine first and second model output data”).
Although Aguilar Alas discloses a dynamic conversational response using ensemble prediction based on a plurality of machine learning models (Fig. 6; ¶0022--¶0023; ¶0086; ¶0098), Aguilar Alas does not specifically disclose comparing the plurality of outputs to the second output to determine whether to use an output of the plurality of outputs or the second output to generate a dynamic conversational response; and generating, at the user interface, the dynamic conversational response during the conversational interaction.
However, Harding discloses generating, at the user interface, the dynamic conversational response during the conversational interaction (col. 2, lines 3-28, “generating replies in response to recognitions automatically using a processor or computer”; col. 7, lines 26-47, “With the words and phrase patterns isolated and classified by the predictions, some chat vectors are processed by a supervised learning algorithm that interprets the text based on a knowledge base at 906 associated with the prediction to automatically generate a response at 908 that is transmitted to the I/O engine 910 of FIG. 9”; col. 9, lines 45-56, “the conversational system ingests input in the form of a JavaScript Object Notation (JSON), parses the input, processes the input via a machine learning algorithm and a grammar-based or natural language recognition engine, scores its determination, compares the determination to an appropriate reply from a knowledge base, and either automatically responds back to the user or routes the request to a legacy chat system for a manual reply. The conversational systems automatically generate replies when the confidence level of the recognized response exceeds a predetermined threshold”; claim 1, “receiving, by a conversational assistant pod within a container device, a real time transmission of a text message from a sender to a receiver over a distributed network through a chat system that enables the receiver to transmit an automatically generated response in real time during a chat session”). It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to modify the system of Aguilar Alas to include generating, at the user interface, the dynamic conversational response during the conversational interaction, as taught by Harding because it would improve the responsiveness of the conversational system by providing an immediate and contextually relevant response during the ongoing interaction (Harding; claim 1).
Zhou, on the hand discloses comparing the plurality of outputs to the second output to determine whether to use an output of the plurality of outputs or the second output to generate a dynamic conversational response (Fig. 1; Fig. 2, S203; Fig. 3, 304-305; ¶0055, “when the final result 314 is generated, the response message outputting module 305 may select the processing result of one thread according to the comparison result between the final result 314 and the predicted complete expression 315”; ¶0095, “comparing a final result with the one or more complete expressions in response to of the final result of speech recognition and selecting a response message as output from the one or more response messages”; ¶0123, “a response message outputting module, configured to compare the final result with the one or more complete expressions in response to the generating the final result of speech recognition and select a response message as output from the one or more response messages”). It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to modify the system of Aguilar Alas to include comparing the plurality of outputs to the second output to determine whether to use an output of the plurality of outputs or the second output to generate a dynamic conversational response, as taught by Zhou because it would increase the reliability of the generated conversational response by allowing the system to select the output having a higher likelihood of accurately representing the user’s intended response (Zhou; Abstract; ¶0026; ¶0040).
As to claim 3, Aguilar Alas discloses the method of claim 2, further comprising: determining a third model of the plurality of separately trained models; and training the third model of the plurality of separately trained models on user interface image data (¶0036, “The models trained during the operations discussed with regard to FIG. 1A may be used during runtime evaluation to determine sentiment information”; ¶0085, “one or more trained machine learning models (such as neural networks, classifiers, etc.)”; ¶0099, “Trained models come in a variety of forms including trained classifiers, Support Vector Machines (SVMs), neural networks (such as deep neural networks (DNNs), recurrent neural networks (RNNs), or convolutional neural networks (CNNs)) and others”; ¶0100, “Models may be trained and operated according to various machine learning techniques”; Claim 3, “using a third machine learning model to determine a likelihood the second utterance corresponds to the first sentiment category”).
As to claim 6, Aguilar Alas discloses the method of claim 2, further comprising: determining a fifth model of the plurality of separately trained models; and training the fifth model of the plurality of separately trained models on transaction data (¶0036, “The models trained during the operations discussed with regard to FIG. 1A may be used during runtime evaluation to determine sentiment information”; ¶0085, “one or more trained machine learning models (such as neural networks, classifiers, etc.)”; ¶0099, “Trained models come in a variety of forms including trained classifiers, Support Vector Machines (SVMs), neural networks (such as deep neural networks (DNNs), recurrent neural networks (RNNs), or convolutional neural networks (CNNs)) and others”; ¶0100, “Models may be trained and operated according to various machine learning techniques”; Claim 3, “using a third machine learning model to determine a likelihood the second utterance corresponds to the first sentiment category”).
As to claim 7, Aguilar Alas discloses the method of claim 2, wherein the second model comprises an ensemble function and wherein the second model that is trained on a second plurality of outputs generated by the plurality of machine learning models (¶0039, “second feature vectors using the trained model to determine first and second model output data. The system(s) 120 may process the first and second model output data using another ML model (e.g., an attention model) to determine third model data”).
As to claim 8, Aguilar Alas discloses the method of claim 2, further comprising: determining an aggregate measure of performance across all possible classification thresholds for the plurality of separately trained models and the second model; determining which model of both (i) the plurality of separately trained models and (ii) the second model has the highest aggregate measure of performance (¶0086, “a number of different machine learning models/techniques such as GMM, neural networks, etc. For example, the confidence component 424 may be a classifier configured to map a score output by the scoring component 422 to a confidence value”; ¶0110, “model 520 may be configured to take in output data from the trained model 515 and produce an output for every time step (e.g., a 10 ms audio frame). The attention model 520 may be configured to aggregate information from different time intervals/audio frames of the input audio data to determine how certain parts of the utterance affects determining of the sentiment”; ¶0118, “obtain the weighted sequence, and the sum of the resulting weighted vectors is z. The output (z) of the attention model 620, 625 may be processed to predict a sentiment category during training. The model weights and parameters learned by the attention model 620 during the training process is stored and used to determine the attention model 520 used during runtime”; ¶0120; ¶0126, “The ML model 630 may be trained separately using the text data 606, and the determined weights and parameters may be stored/frozen while the ML model 615 is trained”).
As to claim 10, Aguilar Alas discloses the method of claim 2, wherein a first output of the plurality of outputs is multi-label, and wherein training data for a respective model of the plurality of separately trained models that generated the first output is multi-class (Fig. 12; ¶0055, “The NLU component 260 may send NLU results data (which may include tagged text data, indicators of intent, etc.) to the orchestrator component 230. The orchestrator component 230 may send the NLU results data to a skill(s) 290”; ¶0099, “Trained models come in a variety of forms including trained classifiers, Support Vector Machines (SVMs), neural networks (such as deep neural networks (DNNs), recurrent neural networks (RNNs), or convolutional neural networks (CNNs)) and others”).
As to claim 11, Aguilar Alas discloses the method of claim 2, wherein the plurality of separately trained models comprises a single multi-class convolutional neural network, wherein a first output of the plurality of separately trained models generated by the single multi-class convolutional neural network is a multi-value vector of probabilities, and wherein each value in the multi-value vector corresponds to a given user intent (Fig. 12; ¶0055, “The NLU component 260 may send NLU results data (which may include tagged text data, indicators of intent, etc.) to the orchestrator component 230. The orchestrator component 230 may send the NLU results data to a skill(s) 290”; ¶0099, “Trained models come in a variety of forms including trained classifiers, Support Vector Machines (SVMs), neural networks (such as deep neural networks (DNNs), recurrent neural networks (RNNs), or convolutional neural networks (CNNs)) and others”).
As to claim 12, Aguilar Alas discloses the method of claim 11, wherein the first output is stacked with the second output by the second model (¶0039, “The system(s) 120 processes the first and second feature vectors using the trained model to determine first and second model output data. The system(s) 120 may process the first and second model output data using another ML model (e.g., an attention model) to determine third model data”).
As to claim 13, Aguilar Alas discloses the method of claim 2, wherein the first feature input indicates an application interface from which a user interface was launched (¶0022; ¶0023).
As to claim 14, it is rejected for the same reasons set forth above with claim 2. In addition, Aguilar Alas discloses wherein determining whether to use an output of the plurality of outputs or the second output to generate the dynamic conversational response further comprises: determining a first measure of performance and a second measure of performance for each of the plurality of separately trained models and the second model based on (i) the plurality of outputs and (ii) the second output; aggregating the first measure of performance and the second measure of performance to generate an aggregated measure of performance for each of the plurality of separately trained models and the second model; comparing the generated aggregated measure of performances to identify a model with the greatest aggregate measure of performance (¶0086, “a number of different machine learning models/techniques such as GMM, neural networks, etc. For example, the confidence component 424 may be a classifier configured to map a score output by the scoring component 422 to a confidence value”; ¶0110, “model 520 may be configured to take in output data from the trained model 515 and produce an output for every time step (e.g., a 10 ms audio frame). The attention model 520 may be configured to aggregate information from different time intervals/audio frames of the input audio data to determine how certain parts of the utterance affects determining of the sentiment”; ¶0118, “obtain the weighted sequence, and the sum of the resulting weighted vectors is z. The output (z) of the attention model 620, 625 may be processed to predict a sentiment category during training. The model weights and parameters learned by the attention model 620 during the training process is stored and used to determine the attention model 520 used during runtime”; ¶0120; ¶0126, “The ML model 630 may be trained separately using the text data 606, and the determined weights and parameters may be stored/frozen while the ML model 615 is trained”).
As to claim 15, it is rejected for the same reasons set forth above with claim 2. In addition, Aguilar Alas discloses the method of claim 2, wherein determining whether to use an output of the plurality of outputs or the second output to generate the dynamic conversational response further comprises: determining a first measure of performance and a second measure of performance for each of the plurality of separately trained models and the second model based on (i) the plurality of outputs and (ii) the second output; aggregating the first measure of performance and the second measure of performance to generate an aggregated measure of performance for each of the plurality of separately trained models and the second model; comparing the generated aggregated measure of performances to identify a model with the greatest aggregate measure of performance (¶0086, “a number of different machine learning models/techniques such as GMM, neural networks, etc. For example, the confidence component 424 may be a classifier configured to map a score output by the scoring component 422 to a confidence value”; ¶0110, “model 520 may be configured to take in output data from the trained model 515 and produce an output for every time step (e.g., a 10 ms audio frame). The attention model 520 may be configured to aggregate information from different time intervals/audio frames of the input audio data to determine how certain parts of the utterance affects determining of the sentiment”; ¶0118, “obtain the weighted sequence, and the sum of the resulting weighted vectors is z. The output (z) of the attention model 620, 625 may be processed to predict a sentiment category during training. The model weights and parameters learned by the attention model 620 during the training process is stored and used to determine the attention model 520 used during runtime”; ¶0120; ¶0126, “The ML model 630 may be trained separately using the text data 606, and the determined weights and parameters may be stored/frozen while the ML model 615 is trained”).
As to claim 16, it is rejected for the same reasons set forth above with claim 2. In addition, Aguilar Alas discloses one or more non-transitory computer-readable media comprising instructions that, when executed by one or more processors (Figs. 14-15), cause operations comprising:
obtaining a set of outputs from a first model based on inputting a first feature input derived from a first user action related to a conversational interaction into the first model (¶0032, “The system(s) 120 receives (134) text data representing the utterance. The system(s) 120 determines (136) at least one lexical feature vector using the text data”; ¶0037, “system(s) 120 receives (150) input audio data representing an utterance”; Claim 4, “wherein receiving the input audio data comprises receiving the second utterance spoken by a first user”), wherein the first model comprises a set of models separately trained on a specific dataset (¶0126, “The ML model 630 may be trained separately using the text data 606, and the determined weights and parameters may be stored/frozen while the ML model 615 is trained”);
inputting a second feature input, based on the set of outputs, into a second model to generate a second output (¶0032, “The system(s) 120 receives (134) text data representing the utterance. The system(s) 120 determines (136) at least one lexical feature vector using the text data. The ASR component 250, as described below, may generate an ASR output including text data representative of an utterance represented in the training audio data”; ¶0038, “The trained model is trained using acoustic and lexical features to detect sentiment in audio data. The model output data may indicate a score or probability of utterance corresponding to one of four sentiment categories (e.g., anger, happiness, sadness, and neutral). The system(s) 120 determines (156), using the model output data, a sentiment category for the utterance”).
As to claims 17-18, they are rejected for the same reasons set forth above with claims 7-8, respectively.
As to claims 19-20, they are rejected for the same reasons set forth above with claims 14-15, respectively.
Claims 1, 4-5 and 9 are rejected under 35 U.S.C. 103 as being unpatentable over Aguilar Alas et al. (US 2021/0104245 A1), Zhou (US 2019/0206397 A1), Harding et al. (US 11,038,821 B1), further in view of Candel et al. (US 2019/0295000 A1).
As to claims 4-5, although Aguilar Alas discloses determining a fourth model of the plurality of separately trained models (¶0126, “The ML model 630 may be trained separately using the text data 606, and the determined weights and parameters may be stored/frozen while the ML model 615 is trained”); and analyzing training data, decision making (¶0031, “The acoustic feature vectors may be derived by spectral analysis of the training audio data”; ¶0034; ¶0050, “Such an architecture may estimate the posteriors of wakewords with context information, either by stacking frames within a context window for DNN, or using RNN. Follow-on posterior threshold tuning or smoothing is applied for decision making”; ¶0117; ¶0144), Aguilar Alas does not specifically disclose the fourth model performs a Weight of Evidence (WOE) analysis.
However, Candel discloses the fourth model performs a Weight of Evidence (WOE) analysis (¶0043; ¶0057, “A weight of evidence may create log-likelihood type of features using the weights of evidence (WOE) transformation method. The weight of evidence tells the predictive power of an independent variable in relation to the dependent variable”; ¶0089; ¶0120). It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to modify the system of Aguilar Alas to include wherein the fourth model performs a Weight of Evidence (WOE) analysis, as taught by Candel because it would improve prediction reliability by quantifying the evidentiary strength of features (Candel; ¶0043; ¶0057; ¶0089; ¶0120).
As to claim 1, it is rejected for the same reasons set forth above with claims 2 and 4-5. In addition, Aguilar Alas discloses
storage circuitry configured to store (¶0035, “stores (144) the combined sentiment detection model as a trained model configured to detect sentiments in audio data”):
a first machine learning model comprising an ensemble of the plurality of machine learning models, wherein each machine learning model of the plurality of machine learning models are respectively trained on user interface image data, transaction data (¶0033, “the first ML model is trained to process acoustic features derived from the training audio data, while the second ML model is trained to process lexical features derived from text data corresponding to the training audio data”; ¶0039, “the system(s) 120 may determine a first portion of the input audio data corresponding to a first word of the utterance and a second portion of the input audio data corresponding to a second word of the utterance”; ¶0115, “The system trains two separate machine learning (ML) models (615 and 630) focused on different modalities. For example, ML model 615 is trained to process acoustic features derived from the training audio data 602, while ML model 630 is trained to process lexical features derived from text data 606”);
a second machine learning model trained on a first plurality of outputs generated by the plurality of machine learning models (¶0038; ¶0039, “the system(s) 120 may determine a first portion of the input audio data corresponding to a first word of the utterance and a second portion of the input audio data corresponding to a second word of the utterance. The system(s) 120 determines a first feature vector using the first portion of input audio data and a second feature vector using the second portion of the input audio data”; -¶0044, “determine time data indicating when the portion of the input audio data is received by the device 110. The output data may include the text data corresponding to the portion of input audio data, the time data”);
control circuitry configured to: receive a first user action related to a conversational interaction with a user interface (¶0032, “The system(s) 120 receives (134) text data representing the utterance. The system(s) 120 determines (136) at least one lexical feature vector using the text data”; ¶0037, “system(s) 120 receives (150) input audio data representing an utterance”; Claim 4, “wherein receiving the input audio data comprises receiving the second utterance spoken by a first user”);
generate the second output by inputting the second feature input into the second machine learning model (¶0033, “trains (140) a second machine learning (ML) model using the lexical feature vector to determine lexical-based-sentiment model data”; ¶0039, “processes the first and second feature vectors using the trained model to determine first and second model output data”); and
input/output circuitry configured to: generate for display, at the user interface, the dynamic conversational response for the conversational interaction (¶0044, “The output data may be displayed on a user device 110a or the device 110b”; ¶0171; ¶0175).
As to claim 9, Candel discloses in response to determining that a respective model of the plurality of separately trained models has a lowest aggregate measure of performance, replacing the respective model with a sixth model (¶0092, “the top performing machine learning models (e.g., top half, top number, top 10%, etc.) of the one or more surviving machine learning models and the one or more evolved machine learning models are retained as surviving machine learning models and the non-top performing machine learning models (e.g., bottom half, all models except for the top two, bottom 90%, etc.) the one or more surviving machine learning models and the one or more evolved machine learning models are discarded or replaced with one or more evolved machine learning models”; ¶0114, “In the event the average of the overall accuracies exceeds the distribution of the entry validation scores of another model, the model is determined to be an improvement of the other model and replaces the other model as a surviving machine learning model”). It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to modify the system of Aguilar Alas to include in response to determining that a respective model of the plurality of separately trained models has a lowest aggregate measure of performance, replacing the respective model with a sixth model, as taught by Candel because it would maximize the overall system accuracy and prevent stagnated performance by continuously filtering out weak links (Candel; ¶0092).
Conclusion
The prior art made of record and not relied upon is considered pertinent to applicant's disclosure.
Ho et al. (US 2020/0074984), Hill et al. (US 2020/0193265), Sait M A et al. (US 2023/0013842), Asthana et al. (US 2021/0097110), Semeniuk et al. (US 11,777,874), Wan et al. (US 2020/0335083), Lee et al. (US 2021/0110259), Rodriguez et al. (US 2018/0367484) disclose suggested items for use with embedded applications in chat conversations.
Any inquiry concerning this communication or earlier communications from the examiner should be directed to JUNGWON CHANG whose telephone number is (571)272-3960. The examiner can normally be reached 9AM-5:30PM.
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, GLENTON BURGESS can be reached at (571)272-3949. 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.
/JUNGWON CHANG/Primary Examiner, Art Unit 2454 September 30, 2026
Prosecution Insights