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 1-9 are pending. Claims 1, 4, and 7 are independent.
Claims 2-3 depend from Claim 1.
Claims 5-6 depend from Claim 4.
Claims 8-9 depend from Claim 7.
This Application was published as U.S. 2025/0201250.
Information Disclosure Statement
The information disclosure statement (IDS) submitted on 11 Sep 2024 and 12 Nov 2025 are in compliance with the provisions of 37 CFR 1.97. Accordingly, the information disclosure statement is being considered by the examiner.
Claim Rejections - 35 USC § 112
The following is a quotation of 35 U.S.C. 112(b):
(b) CONCLUSION.—The specification shall conclude with one or more claims particularly pointing out and distinctly claiming the subject matter which the inventor or a joint inventor regards as the invention.
The following is a quotation of 35 U.S.C. 112 (pre-AIA ), second paragraph:
The specification shall conclude with one or more claims particularly pointing out and distinctly claiming the subject matter which the applicant regards as his invention.
Claims 1-3, 5-6, and 8-9 rejected under 35 U.S.C. 112(b) or 35 U.S.C. 112 (pre-AIA ), second paragraph, as being indefinite for failing to particularly point out and distinctly claim the subject matter which the inventor or a joint inventor (or for applications subject to pre-AIA 35 U.S.C. 112, the applicant), regards as the invention.
Claim 1 recites the limitation “the learning model classification means” in line 12 and the “the condition classification means” in line 13. There is insufficient antecedent basis for these limitations in the claim because references to the learning model classification means and the condition classification means appears to have been deleted from the amended claim.
For the sake of examination, it will be assumed that the learning model classification means is a machine learning model that outputs a classification result as discussed in claim 1 lines 4-7. For the sake of examination, it will be assumed that the condition classification means is some mechanism that generates the classification result as claimed in claim 1 lines 8-10.
Claim 2 recites the limitation “the specified identifier” in lines 7-8. There is insufficient antecedent basis for this limitation in the claim because the claim refers to the “identifier given to the sound data” and the “the specified identifier” refers to “information registered in advance for each identifier.” Claim 1, from which Claim 2 depends, does not refer to any identifier, so it is not clear what the “the specified identifier” is referring to when there is more than one identifier because each identifier has information registered in advance.
For the sake of examination, it will be assumed that the “specified identifier” is the information about the identifier of a speaker that is given to the classification mechanism for the purpose of classification of the sound data. In addition, extracting information corresponding to the specified identifier will be interpreted as extracting any information about the identifier.
Claim 3 recites the limitation “the learning model classification means” in line 9 and the “the condition classification means” in line 9-10. There is insufficient antecedent basis for these limitations in the claim. The “the learning model classification means” seems to have been deleted from the amended claims; however, there is no limitation in reference to the “condition classification means” because the sound classification means was deleted from the amended claims.
For the sake of examination, it will be assumed that the learning model classification means is a machine learning model that outputs a classification result as discussed in claim 3 lines 6-7. For the sake of examination, it will be assumed that the condition classification means is some mechanism that generates the classification result as was stated in claim 1 lines 8-10.
Claim 5 recites the limitation “the specified identifier” in lines 5-6 and 8. There is insufficient antecedent basis for this limitation in the claim because the claim refers to the “identifier given to the sound data” and the “the specified identifier” refers to “information registered in advance for each identifier.” Claim 4, from which Claim 5 depends, does not refer to any identifier, so it is not clear what the “the specified identifier” is referring to when there is more than one identifier because each identifier has information registered in advance.
For the sake of examination, it will be assumed that the “specified identifier” is the information about the identifier of a speaker that is given to the classification mechanism for the purpose of classification of the sound data. In addition, extracting information corresponding to the specified identifier will be interpreted as extracting any information about the identifier.
Claim 5 lines 2-3 refers to “an identifier of a speaker is given to the sound data to be classified” and in lines 4-5 refers to “the identifier given to the sound data is specified from the sound data to be classified.” Thus, the identifier is “given to the sound data to be classified” and is “specified from the sound data to be classified.” It is not clear how the identifier can be both given to the sound data and specified from the sound data to be classified.
For the sake of examination, it will be assumed that the identifier given to the sound data is the same identifier specified from the sound data.
Claim 6 is rejected because it is a dependent claim of claim 5 which was rejected under 35 U.S.C. 112(b).
Claim 8 recites the limitation “the specified identifier” in lines 6 and 8. There is insufficient antecedent basis for this limitation in the claim because the claim refers to the “identifier given to the sound data” and the “the specified identifier” refers to “information registered in advance for each identifier.” Claim 7, from which Claim 8 depends, does not refer to any identifier, so it is not clear what the “the specified identifier” is referring to when there is more than one identifier because each identifier has information registered in advance.
For the sake of examination, it will be assumed that the “specified identifier” is the information about the identifier of a speaker that is given to the classification mechanism for the purpose of classification of the sound data. In addition, extracting information corresponding to the specified identifier will be interpreted as extracting any information about the identifier.
Claim 9 is rejected because it is a dependent claim of claim 8 which was rejected under 35 U.S.C. 112(b).
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 1-9 are rejected under 35 U.S.C. 103 as being unpatentable over Tang (US2021/0217437 hereinafter Tang) in view of Ando et al. (US2021/0166679 hereinafter Ando) and Huang et al.(US2018/0293988 hereinafter Huang)
With regards to claim 1, Tang teaches:
A sound classification apparatus comprising: at least one memory storing instructions; and [Tang Fig 5 teaches memory (502) stores instructions (Par [0081])]
at least one processor configured to execute the instructions to: [Tang Fig 5 teaches processor (501) configured to execute the instructions (Par [0080])]
input a sound data to be classified into a machine learning model [Tang Fig 1 step S101 teaches receiving a sound data and step S102 teaches classifying the data by inputting the “user audio received in S101 into the audio classification model” (Par [0028])]
generated by machine learning using data that serve as training data, and [Tang teaches “audio classification model may be a classification model obtained by training based on the machine learning algorithm” (Par [0028])]
output a classification result using an output result from the machine learning model; [Tang teaches “audio classification model may output the audio type information based on the input audio information.” (Par [0028])]
With regards to claim 1, Tang fails to teach:
sound data and teacher data that serve as training data, and
With regards to claim 1, Ando teaches:
sound data and teacher data that serve as training data, and [Ando Figures 3-4 teaches model learning apparatus (1) which includes “utterance-with-teacher label storage 10a” (Par [0036, Fig 3-4]) which are sound and teacher data using for training data.
It would be obvious to one of ordinary skill in the art at the time of applicant’s filing to combine the machine learning model for sound classification as taught by Tang with the training a model method of using sound and teacher data as taught by Ando. The motivation to combine the teachings of Tang with Ando is because Ando teaches “utterance of a human is associated with a teacher label of paralinguistic information for classifying the utterance” (Par [0039]) which increase the capabilities of the invention of Tang to learn paralinguistic information of the utterance]
With regards to claim 1, Tang in view of Ando fails to teach:
classify the sound data to be classified, based on information registered in advance, and outputting a classification result; and
classify the sound data to be classified, based on the classification result of the learning model classification means and the classification result of the condition classification means.
With regards to claim 1, Huang teaches:
classify the sound data to be classified, based on information registered in advance, and outputting a classification result; and [Huang Fig 3 teaches classifying the audio signal or sound data by inputting into the “context prediction unit 308 that compares pre-stored context audio signal data of pre-stored context data from a context database 310 to the audio input signal to classify the audio input signal as one of the contexts” (Par [0048]) where pre-stored data is information registered in advance]
classify the sound data to be classified, based on the classification result of the learning model classification means and the classification result of the condition classification means. [Huang Fig 3 teaches speaker score unit (302) that includes the speaker model that creates a “speaker score is also provided to a speaker identity prediction unit 304 that performs the comparison of speaker score to threshold to make a recognition decision” (Par [0047]) were a recognition decision is a classification and the classification result of the condition classification means is used to determine the “threshold obtained from a threshold generation unit 306 to provide a recognition decision.” (Par [0043])
It would be obvious to one of ordinary skill in the art at the time of applicant’s filing to combine the machine learning model for sound classification as taught by Tang in view of Ando with the context classification method of speaker recognition as taught by Huang. The motivation to combine the teachings of Tang and Ando with Huang is because Huang teaches
“context database provides data for a continuous learning model thereby increasing the accuracy of the confidence model's error estimation” (Par [0045]) which increase the capabilities of the invention of Tang and Ando to adapt to the current environment to provide better classification of data]
With regards to claim 2, Tang in view of Ando and Huang teaches:
All the limitations of claim 1
wherein the sound data is voice data, [Tang Fig 1 teaches “user audio may be a piece of voice uttered by the user casually speaking or singing, or the voice uttered by the user reading aloud a preset text, or the voice uttered by the user singing a preset lyric, and so on” (Par [0021])]
and an identifier of a speaker is given to the sound data to be classified, and
the one or more processors further specifies, from the sound data to be classified, the identifier given to the sound data, refers to the specified identifier in information registered in advance for each identifier, extracts the information corresponding to the specified identifier, and outputs the extracted information as the classification result. [Tang Fig 1 step S102 obtains audio type information, for example “gender and voice category” (Par [0024]) to be classified and uses various methods to such as “ the user audio may be input into a voice gender classification model obtained by training based on a machine learning algorithm, to obtain the gender of the user audio … [and] the executing body may use the gender and the voice category of the user audio as the audio type information of the user audio.” (Par [0025])]
With regards to claim 3, Tang in view of Ando and Huang teaches:
All the limitations of claim 2
wherein the machine learning model is generated by machine learning using voice
data and information characterizing voice, the one or more processors further; outputs information characterizing voice, corresponding to the sound data to be classified, as the classification result, and [Tang teaches “the user audio may be input into a voice gender classification model obtained by training based on a machine learning algorithm, to obtain the gender of the user audio. Here, the voice gender classification model may be obtained by training based on a large amount of training data, and is used to predict the gender of a speaker corresponding to the voice based on an input voice.” (Par [0025])]
outputs information in which the classification result of the learning model classification means and the classification result of the condition classification means are combined, as a result of the classification. [Huang Fig 3 teaches speaker score unit (302) that includes the speaker model that creates a “speaker score is also provided to a speaker identity prediction unit 304 that performs the comparison of speaker score to threshold to make a recognition decision” (Par [0047]) were a recognition decision is a classification and the classification result of the condition classification means is used to determine the “threshold obtained from a threshold generation unit 306 to provide a recognition decision.” (Par [0043])]
With regards to claim 4, Tang teaches:
A sound classification method comprising: inputting sound data to be classified into a machine learning model [Tang Fig 1 step S101 teaches receiving a sound data and step S102 teaches classifying the data by inputting the “user audio received in S101 into the audio classification model” (Par [0028])]
generated by machine learning using data that serve as training data, and [Tang teaches “audio classification model may be a classification model obtained by training based on the machine learning algorithm” (Par [0028])]
outputting a classification result using an output result from the machine learning model; [Tang teaches “audio classification model may output the audio type information based on the input audio information.” (Par [0028])]
With regards to claim 4, Tang fails to teach:
sound data and teacher data that serve as training data, and
With regards to claim 4, Ando teaches:
sound data and teacher data that serve as training data, and [Ando Figures 3-4 teaches model learning apparatus (1) which includes “utterance-with-teacher label storage 10a” (Par [0036, Fig 3-4]) which are sound and teacher data using for training data.
It would be obvious to one of ordinary skill in the art at the time of applicant’s filing to combine the machine learning model for sound classification as taught by Tang with the training a model method of using sound and teacher data as taught by Ando. The motivation to combine the teachings of Tang with Ando is because Ando teaches “utterance of a human is associated with a teacher label of paralinguistic information for classifying the utterance” (Par [0039]) which increase the capabilities of the invention of Tang to learn paralinguistic information of the utterance]
With regards to claim 4, Tang in view of Ando fails to teach:
classifying the sound data to be classified, based on information registered in
advance, and outputting a classification result; and
classifying the sound data to be classified, based on the classification result of
machine learning model and the classification result using the information.
With regards to claim 4, Huang teaches:
classifying the sound data to be classified, based on information registered in
advance, and outputting a classification result; and [Huang Fig 3 teaches classifying the audio signal or sound data by inputting into the “context prediction unit 308 that compares pre-stored context audio signal data of pre-stored context data from a context database 310 to the audio input signal to classify the audio input signal as one of the contexts” (Par [0048]) where pre-stored data is information registered in advance]
classifying the sound data to be classified, based on the classification result of
machine learning model and the classification result using the information. [Huang Fig 3 teaches speaker score unit (302) that includes the speaker model that creates a “speaker score is also provided to a speaker identity prediction unit 304 that performs the comparison of speaker score to threshold to make a recognition decision” (Par [0047]) were a recognition decision is a classification and the classification result of the condition classification means is used to determine the “threshold obtained from a threshold generation unit 306 to provide a recognition decision.” (Par [0043])
It would be obvious to one of ordinary skill in the art at the time of applicant’s filing to combine the machine learning model for sound classification as taught by Tang in view of Ando with the context classification method of speaker recognition as taught by Huang. The motivation to combine the teachings of Tang and Ando with Huang is because Huang teaches
“context database provides data for a continuous learning model thereby increasing the accuracy of the confidence model's error estimation” (Par [0045]) which increase the capabilities of the invention of Tang and Ando to adapt to the current environment to provide better classification of data]
Claim 5 is a method claim with limitations corresponding to the limitations of apparatus Claim 2 and is rejected under similar rationale.
Claim 6 is a method claim with limitations corresponding to the limitations of apparatus Claim 3 and is rejected under similar rationale.
With regards to claim 7, Tang teaches:
A non-transitory computer-readable recording medium including a program recorded thereon, the program including instruction that cause a computer to carry out: [Tang Fig 5 teaches memory (502) stores instructions (Par [0081])]
inputting sound data to be classified into a machine learning model [Tang Fig 1 step S101 teaches receiving a sound data and step S102 teaches classifying the data by inputting the “user audio received in S101 into the audio classification model” (Par [0028])]
generated by machine learning using data that serve as training data, and [Tang teaches “audio classification model may be a classification model obtained by training based on the machine learning algorithm” (Par [0028])]
outputting a classification result using an output result from the machine learning model; [Tang teaches “audio classification model may output the audio type information based on the input audio information.” (Par [0028])]
With regards to claim 7, Tang fails to teach:
sound data and teacher data that serve as training data, and
With regards to claim 7, Ando teaches:
sound data and teacher data that serve as training data, and [Ando Figures 3-4 teaches model learning apparatus (1) which includes “utterance-with-teacher label storage 10a” (Par [0036, Fig 3-4]) which are sound and teacher data using for training data.
It would be obvious to one of ordinary skill in the art at the time of applicant’s filing to combine the machine learning model for sound classification as taught by Tang with the training a model method of using sound and teacher data as taught by Ando. The motivation to combine the teachings of Tang with Ando is because Ando teaches “utterance of a human is associated with a teacher label of paralinguistic information for classifying the utterance” (Par [0039]) which increase the capabilities of the invention of Tang to learn paralinguistic information of the utterance]
With regards to claim 7, Tang in view of Ando fails to teach:
classifying the sound data to be classified, based on information registered in
advance, and outputting a classification result; and
classifying the sound data to be classified, based on the classification result of
machine learning model and the classification result using the information.
With regards to claim 7, Huang teaches:
classifying the sound data to be classified, based on information registered in
advance, and outputting a classification result; and [Huang Fig 3 teaches classifying the audio signal or sound data by inputting into the “context prediction unit 308 that compares pre-stored context audio signal data of pre-stored context data from a context database 310 to the audio input signal to classify the audio input signal as one of the contexts” (Par [0048]) where pre-stored data is information registered in advance]
classifying the sound data to be classified, based on the classification result of
machine learning model and the classification result using the information. [Huang Fig 3 teaches speaker score unit (302) that includes the speaker model that creates a “speaker score is also provided to a speaker identity prediction unit 304 that performs the comparison of speaker score to threshold to make a recognition decision” (Par [0047]) were a recognition decision is a classification and the classification result of the condition classification means is used to determine the “threshold obtained from a threshold generation unit 306 to provide a recognition decision.” (Par [0043])
It would be obvious to one of ordinary skill in the art at the time of applicant’s filing to combine the machine learning model for sound classification as taught by Tang in view of Ando with the context classification method of speaker recognition as taught by Huang. The motivation to combine the teachings of Tang and Ando with Huang is because Huang teaches
“context database provides data for a continuous learning model thereby increasing the accuracy of the confidence model's error estimation” (Par [0045]) which increase the capabilities of the invention of Tang and Ando to adapt to the current environment to provide better classification of data]
Claim 8 is a non-transitory computer-readable recording medium claim with limitations corresponding to the limitations of apparatus Claim 2 and is rejected under similar rationale.
Claim 9 is a non-transitory computer-readable recording medium claim with limitations corresponding to the limitations of apparatus Claim 3 and is rejected under similar rationale.
Conclusion
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