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 .
Response to Amendment
In response to the amendment filed 7/14/2025; claims 1-6, 9-19 and 21-22 are pending; claims 7 – 8 and 20 have been cancelled.
Claim Rejections - 35 USC § 101
35 U.S.C. 101 reads as follows:
Whoever invents or discovers any new and useful process, machine, manufacture, or composition of matter, or any new and useful improvement thereof, may obtain a patent therefor, subject to the conditions and requirements of this title.
Claims 1-6, 9-19 and 21-22 are rejected under 35 U.S.C. 101 because the claimed invention is directed to a judicial exception (i.e., a law of nature, a natural phenomenon, or an abstract idea) without significantly more.
Step 1: Is the claimed invention a statutory category of invention?
Claims 1, 9 and 17 are directed to a method for generating reading competency score (Step 1, Yes).
Step 2A, Prong 1: Does the claim recite an abstract idea?
The limitation of steps:
… outputting a text for display at a reading interface; receiving audio data of the text being read orally by a student; converting the audio data to converted text by a speech-to-text engine; identifying an error in the converted text by detecting a difference between the converted text and the text; classifying, with a trained machine classifier, the error into an error category, wherein the error category is a self-correction; receiving feedback from a user indicating that the error category is incorrect; using the feedback to generate a misclassification record; and retraining the machine classifier using data from the misclassification record as training data, wherein the training data from the misclassification record comprises the audio data of the text being read orally by the student; generating a reading competency score using the error category; and outputting the reading competency score for display at a reading-analysis interface as drafted, is a process that, under its broadest reasonable interpretation, covers performance of the limitation in the mind but for the recitation of generic computer components. This type of mental process can be practically performed in the human mind, for instance by a human teach mentally performs each of the claimed steps. Thus, the claim recites a mental process (Step 2A, Prong 1: yes).
Step 2A, Prong 2: Does the claim recite additional elements that integrate the judicial exception into a practical application?
Per the 2019 Revised Patent Subject Matter Eligibility Guidance, if a claim as a whole integrates the recited judicial exception into a practical application of that exception, a claim is not "directed to" a judicial exception. Alternatively, a claim that does not integrate a recited judicial exception into a practical application is directed to the exception. Evaluating whether a claim integrates an abstract idea into a practical application is performed by a) identifying whether there are any additional elements recited in the claim beyond the abstract idea, and b) evaluating those additional elements individual and in combination to determine whether they integrate the abstract idea into a practical application, using one or more of the considerations laid out by the Supreme Court and the Federal Circuit. Exemplary considerations indicative that an additional element (or combination of elements) may have or has not been integrated into a practical application are set forth in the 2019 PEG.
With respect to the instant claims, claims 1, 9 , 17 recite the additional elements of: a computing device; a reading interface, a reading-analysis interface, It is particularly noted that the use of a computing device "as a tool" to perform an abstract method and steps that only amount to extra solution activity are indicated in the 2019 PEG as examples that an additional element has not been integrated into a practical application. Even in combination, the recited additional elements do not integrate the abstract idea into a practical application because they do not impose any meaningful limits, such as an improvement to a computing system, on practicing the abstract idea (STEP 2A, Prong 2: NO).
Step 2B: Does the claim recite additional elements that amount to significantly more than the judicial exception?
Claims 1, 9 and 17 recite the additional elements of: a computing device; a reading interface, a reading-analysis interface set forth above for Step 2A, Prong 2. Regarding these limitations: Applicant's specification only describes these features in a highly generic manner by stating that " Various functions described herein as being performed by one or more entities may be carried out by hardware, firmware, and/or software. For instance, some functions may be carried out by a processor executing instructions stored in memory" in the Applicant’s specification, page. 6, para. [00029]. Claims 9 and 17 do not require any statutory product to be tied to the claimed method. There is no indication in the Specification that Applicants have achieved an advancement or improvement computing technology. Dependent claims 2 – 6, 10 – 16, 18 – 19 and 21 - 22 inherit the deficiencies of their respective parent claims through their dependencies and do not recite additional limitations sufficient to direct the claims to more than the claimed abstract idea, and are thus rejected for the same reasons.
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.
The factual inquiries for establishing a background for determining obviousness under 35 U.S.C. 103 are summarized as follows:
1. Determining the scope and contents of the prior art.
2. Ascertaining the differences between the prior art and the claims at issue.
3. Resolving the level of ordinary skill in the pertinent art.
4. Considering objective evidence present in the application indicating obviousness or nonobviousness.
Claims 1 - 2, 9, 12, 17, 21 – 22 are rejected under 35 U.S.C. 103 as being unpatentable over Wasowicz et al. (US 2002/0164563 A1) in view of Meng et al. (US 2020/0334538 A1) and Wilson et al. (US 2018/0144269 A1).
Re claims 1, 9:
1. Wasowicz teaches [O]ne or more non-transitory computer storage media comprising computer-executable instructions that when executed by a computing device cause the computing device to perform a method of reading instruction (Wasowicz, Abstract) comprising:
outputting a text for display at a reading interface (Wasowicz, fig. 23; [0127], “the module may present a spoken word and a picture of the item in step 280 and query the user about which item in a sequence of items”; [0130]);
receiving audio data of the text being read orally by a student (Wasowicz, fig. 23);
converting the audio data to converted text by a speech-to-text engine (Wasowicz, [0111]);
identifying an error in the converted text by detecting a difference between the converted text and the text (Wasowicz, [0111], “speak the name of each item into a microphone that is interpreted by the speech recognition software in the client computer, transmitted to the server and compared to a correct response by the speech recognition software in the server so that the scorer may determine whether or not the child correctly identified each item”; [0126]; [0013], [0127], “If the response is incorrect, the module may determine the number of consecutive errors for the particular ending sound in step 288, compare the calculated number to a predetermined number in step 290 and display a next word”);
classifying, with a trained machine classifier, the error into an error category (Wasowicz, fig. 29, 740; [0056], “describe the different sounds units types, syllable types and phoneme types that may be tested using the diagnostic system since these types of sound units, syllables and phonemes are similar to the types of sound units, syllables and phonemes used in the training tools”; [0098], “indexes are then incremented as described below to analyze each incorrect response for each subtest wherein each incorrect response is compared to each error measure to determine the type of error”; [0099]);
generating a reading competency score using the corrected error category (Wasowicz, [0097], “example, two of the incorrect responses indicate the same type of error (for example, an open syllable rime error) and one indicates a different type of error (for example, a r-controlled vowel rime). In this manner, the data about the particular incorrect responses by the user stored in the database are mapped into the types of errors that are shown by the particular incorrect answer”; [0098], “indexes are then incremented as described below to analyze each incorrect response for each subtest wherein each incorrect response is compared to each error measure to determine the type of error”; [0099]); and
outputting the reading competency score for display at the reading-analysis interface (Wasowicz, fig. 25; [0133] – [0136]; [0141]; figs. 27 – 28; [0144]).
9. One or more non-transitory computer storage media comprising computer-executable instructions that when executed by a computing device cause the computing device to perform a method of reading instruction (Wasowicz, Abstract) comprising:
outputting for display a text from a reading assignment at a reading interface (Wasowicz, fig. 23; [0127], “the module may present a spoken word and a picture of the item in step 280 and query the user about which item in a sequence of items”; [0130]);
receiving audio data of the text being read orally by a student (Wasowicz, fig. 23);
converting, using a speech-to-text engine, the audio data to converted text (Wasowicz, [0111]);
identifying an error in the converted text by detecting a difference between the converted text and the text (Wasowicz, [0111], “speak the name of each item into a microphone that is interpreted by the speech recognition software in the client computer, transmitted to the server and compared to a correct response by the speech recognition software in the server so that the scorer may determine whether or not the child correctly identified each item”; [0126]; [0013], [0127], “If the response is incorrect, the module may determine the number of consecutive errors for the particular ending sound in step 288, compare the calculated number to a predetermined number in step 290 and display a next word”);
classifying, with a trained machine classifier, the error into an error category (Wasowicz, fig. 29, 740; [0056], “describe the different sounds units types, syllable types and phoneme types that may be tested using the diagnostic system since these types of sound units, syllables and phonemes are similar to the types of sound units, syllables and phonemes used in the training tools”; [0098], “indexes are then incremented as described below to analyze each incorrect response for each subtest wherein each incorrect response is compared to each error measure to determine the type of error”; [0099]);
generating a reading competency score using the corrected error category; and
outputting an analysis of the converted text showing a reading error made by the student at the reading-analysis interface (Wasowicz, fig. 25; [0133] – [0136]; [0141]; figs. 27 – 28; [0144]).
Wasowicz does not explicitly disclose misclassification of the error category.
Meng et al. (US 2020/0334538 A1) teaches conditional teacher-student model training. Meng further teaches receiving feedback from a user indicating that the error category is incorrect; using the feedback to generate a misclassification record; and retraining the machine classifier using data from the misclassification record as training data, wherein the training data from the misclassification record comprises the audio data of the text being read orally by the student (Meng, [0018], “incorrectly identify frames in an utterance, which can cause the performance of a student model to degrade”; [0041], “The student model 260 is trained under the supervision of the teacher model 250, wherein each model 250, 260 receives utterances in their respective domains in parallel … The parallel utterances are received by the respective teacher model 250 or student model 260, which may each correctly or incorrectly recognize the utterance as containing a given word”; [0062], “various machine learning techniques may be used to update the student model 260”; [0033], “Automatic Speech Recognition (“ASR”) task where the goal is to learn a student acoustic model”). Therefore, in view of Meng, 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 method/system, by updating the student ASR model as taught by Meng, in order to improve the student
model performance and accuracy (Meng, [0022]; [0047]).
Wilson et al. (US 2018/0144269 A1) teaches a field of machine learning and information retrieval.
displaying, in a reading-analysis interface, a list of candidate error categories associated with the identified error (Wilson, [0072], “FIG. 12, a first classifier ("Classifier 1") can be trained using features of the training data content items based on Attribute 1 and Attribute 2. A second classifier ("Classifier 2") can be trained using features based on a different subset of the attributes. For example, the second classifier can be trained using only features based on Attribute 1”);
receiving a user selection of a corrected error category from among the candidate error categories (Wilson, [0070], “The label assigned to content items 51 and 100 are correct, while the label predicted by the classifier for content item 99”);
using the feedback to generate a misclassification record using the selected corrected error category as a label for the identified error (Wilson, [0079], “content items 3, 5, 7 and 8 are classified incorrectly”);
in response to the user selection of the corrected error category, initiating retraining of the machine classifier using data from the misclassification record as training data to generate a student-specific machine classifier for the student, wherein the training data from the misclassification record comprises the audio data of the text being read orally by the student as input and the selected error category as the label (Wilson, [0070], “The label assigned to content items 51 and 100 are correct, while the label predicted by the classifier for content item 99 is incorrect”; col. 11, claim 7, “re-training the machine-learning classifier based on the number of correct classifications and the number of incorrect classifications”)
Therefore, in view of Wilson, 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 method described in Wasowicz, by retraining the ml as taught by Wilson, since the machine-learning classifier could be trained using both the training data and the validation data sets in the event that the classifier based only on the training data set did not achieve the requisite level of accuracy (Wilson, [0070]).
Re claim 2:
2. The media of claim 1, wherein the method further comprises generating an error report that displays the error and the error category (Wasowicz, fig. 29, 740; [0056], “describe the different sounds units types, syllable types and phoneme types that may be tested using the diagnostic system since these types of sound units, syllables and phonemes are similar to the types of sound units, syllables and phonemes used in the training tools”; [0098], “indexes are then incremented as described below to analyze each incorrect response for each subtest wherein each incorrect response is compared to each error measure to determine the type of error”; [0099]; [0096], “The recommender 108 may use the scores and statistical information generated by the scorer, if requested by the user of the client computer, to recommend one or more training tools that may be used by the child taking the tests on the particular client computer in order to improve the child's ability in any deficient areas”).
Re claim 12:
12. The media of claim 9, wherein the method further comprises: outputting for display a phonemes detail view that identifies phonemes assigned to sounds within the audio data (Wasowicz, fig. 29, 740; [0056], “describe the different sounds units types, syllable types and phoneme types that may be tested using the diagnostic system since these types of sound units, syllables and phonemes are similar to the types of sound units, syllables and phonemes used in the training tools”; [0098], “indexes are then incremented as described below to analyze each incorrect response for each subtest wherein each incorrect response is compared to each error measure to determine the type of error”; [0099]; [0096], “The recommender 108 may use the scores and statistical information generated by the scorer, if requested by the user of the client computer, to recommend one or more training tools that may be used by the child taking the tests on the particular client computer in order to improve the child's ability in any deficient areas”).
Re claim 21:
21. The media of claim 1, wherein the training data used from the misclassification record to retrain the machine classifier further comprises a correct classification as a training label included with the audio data of the text being read orally by the student (Meng, [0018], “incorrectly identify frames in an utterance, which can cause the performance of a student model to degrade”; [0041], “The student model 260 is trained under the supervision of the teacher model 250, wherein each model 250, 260 receives utterances in their respective domains in parallel … The parallel utterances are received by the respective teacher model 250 or student model 260, which may each correctly or incorrectly recognize the utterance as containing a given word”; [0062], “various machine learning techniques may be used to update the student model 260”; [0033], “Automatic Speech Recognition (“ASR”) task where the goal is to learn a student acoustic model”).
Claims 3 – 6 and 15 – 16 are rejected under 35 U.S.C. 103 as being unpatentable over Wasowicz and Meng and Wilson et al. (US 2018/0144269 A1) as applied to claims 1 and 9 above, and further in view of Alison (“How To Analyze Running Records (And Get a Ton of Valuable Information About Your Beginning Readers!)” by Alison, https://learningattheprimarypond.com/blog/how-to-analyze-running-records/, retrieved from Internet Wayback machine, 11/19/2017).
Re claims 3 – 6, 15 - 16:
Wasowicz does not explicitly disclose the reading competency score is a meaning, structural, and visual cues (MSV) score; nor disclose the error category is an attempt, self-correction, or appeal.
Alison teaches a method for analyzing running records for a reader. Alison further teaches: 6. The media of claim 1, wherein the reading competency score is a meaning, structural, and visual cues (MSV) score. 3. The media of claim 1, wherein the method further comprises generating an error report that displays the error and the error category. 3. The media of claim 1, wherein the error category is an appeal. 4. The media of claim 1, wherein the error category is an attempt. 5. The media of claim 1, wherein the error category is a self-correction. 15. The media of claim 9, wherein the error category is an attempt. 16. The media of claim 9, wherein the error category is a self-correction (Alison, pg. 5, “Cueing Systems (M-S-V)”; pg. 7, “… Appealing for help … Rerunning (going back and rereading during tricky parts … Self correcting”). Therefore, in view of Alison, 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 media/method described in Wasowicz, by providing the MSV and reading habit (i.e., self correcting) as taught by Alison, in order to monitor a reader’s word reading habits and general reading behaviors (Alison, pg. 7). A reader is self-monitoring to make sure that what he/she reads looks right, sounds right, and makes sense … self-corrects is likely doing a really good job of monitoring (Alison, pg. 10).
Claims 10 – 11 and 13 – 14 are rejected under 35 U.S.C. 103 as being unpatentable over Wasowicz and Meng and Wilson et al. (US 2018/0144269 A1) as applied to claim 1 above, and further in view of in view of D'Helon et al. (US 2019/0362643 A1) and Dhamija et al. (US 2021/0390492 A1).
Re claim 10
Wasowicz teaches receiving feedback through a parent interface indicating that the error category is incorrect (Wasowicz, [0095]; [0143]). Wasowicz does not explicitly disclose receiving a confirmation through a teacher interface; nor disclose misclassification of the error category.
D’Helon teaches a method, system, and computer program product for improving performance of dialogue-based tutors. D’Helon teaches 10. The media of claim 9, wherein the method further comprises: receiving a confirmation through a teacher interface that the error category is incorrect (D’Helon, [0045], “active reviewers may see the feedback annotation: “My answer is entirely correct” plus (i) the tutor question (ii) the reference answer and (iii) the student answer. Reviewers can either agree or disagree with the annotation”). Therefore, in view of D’Helon, 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 method/system, by providing the machine learning classification described as taught by D’Helon, in order to allow a reviewer to approve or disapprove the feedback candidates (D’Helon, [0006]).
Dhamija teaches systems and techniques are disclosed for a collaboration assessment system that applies machine learning to automatically and objectively evaluate a collaboration effort of two or more individuals (Dhamija, Abstract). Dhamija further teaches in response to the confirmation, generating a misclassification record; and retraining the machine classifier using data from the misclassification record as training data (Dhamija, [0070], “Level A" Classification Codes 114 includes (1) "Effective," (2) "Satisfactory," (3) "Progressive," (4) "Needs Improvement," and (5) "Working Independently," … Level A module 112 of system 100 incorrectly assigns a Level A classification code of "Working Independently" onto video input data 102 that actually depicts an "Effective" … that the classification was incorrect, but also that the classification was incorrect by a full four categories, causing the appropriate machine learning model(s) involved to substantially modify the predictive process”; [0062], “The models of system 100 may then be re-trained with the "augmented" mixup training data ( or "pseudo-data")”; [0106], “the system 100 selects the Level B2 classification code for three of the individuals (e.g., Students 1, 3, and 4)”; [0081], “the training-data annotators may be instructed to assign both a “primary” Level C classification code and a “secondary” Level C classification code to each individual”). Therefore, in view of Dhamija, 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 method/system, by providing the machine learning classification described as taught by Dhamija, in order to augment a set of training data for the machine learning models by intelligently generating additional "pseudo-data" for under-represented categories or classification codes of the training data (Dhamija, [0005] – [0006]).
Re claim 11:
11. The media of claim 10, wherein the method further comprises: in response to the feedback, outputting a feedback notification in the teacher interface (D’Helon, [0045], “active reviewers may see the feedback annotation: “My answer is entirely correct” plus (i) the tutor question (ii) the reference answer and (iii) the student answer. Reviewers can either agree or disagree with the annotation”).
Re claims 13, 14:
13. The media of claim 10, wherein the method further comprises: outputting a classification confidence generated by the speech-to-text engine for a specific phoneme (Wasowicz, fig. 29, 740; [0056], “describe the different sounds units types, syllable types and phoneme types that may be tested using the diagnostic system since these types of sound units, syllables and phonemes are similar to the types of sound units, syllables and phonemes used in the training tools”; [0098], “indexes are then incremented as described below to analyze each incorrect response for each subtest wherein each incorrect response is compared to each error measure to determine the type of error”; [0099]).
14. The media of claim 13, wherein the method further comprises: in response to the classification confidence being less than a threshold (Wasowicz, [0133]; [0136]; Dhamija, [0048], “relative magnitude or confidence value for the annotation”), adding the specific phoneme to a development list for the student; and using the development list to generate a reading-assignment recommendation for an assignment that includes the specific phoneme (Wasowicz, fig. 29, 740; [0056], “describe the different sounds units types, syllable types and phoneme types that may be tested using the diagnostic system since these types of sound units, syllables and phonemes are similar to the types of sound units, syllables and phonemes used in the training tools”; [0098], “indexes are then incremented as described below to analyze each incorrect response for each subtest wherein each incorrect response is compared to each error measure to determine the type of error”; [0099]; [0096], “The recommender 108 may use the scores and statistical information generated by the scorer, if requested by the user of the client computer, to recommend one or more training tools that may be used by the child taking the tests on the particular client computer in order to improve the child's ability in any deficient areas”).
Claims 17 – 19 and 22 are rejected under 35 U.S.C. 103 as being unpatentable over Wasowicz et al. (US 2002/0164563 A1) in view of Meng et al. (US 2020/0334538 A1) and Alison (“How To Analyze Running Records (And Get a Ton of Valuable Information About Your Beginning Readers!)” by Alison, https://learningattheprimarypond.com/blog/how-to-analyze-running-records/, retrieved from Internet Wayback machine, 11/19/2017) and Wilson et al. (US 2018/0144269 A1).
Re claim 17:
17. Wasowicz teaches One or more non-transitory computer storage media comprising computer-executable instructions that when executed by a computing device cause the computing device to perform method of reading instruction (Wasowicz, Abstract) comprising:
outputting for display a text from a reading assignment at a reading interface (Wasowicz, fig. 23; [0127], “the module may present a spoken word and a picture of the item in step 280 and query the user about which item in a sequence of items”; [0130]);
receiving audio data of the text being read orally by a student (Wasowicz, fig. 23);
converting, using a speech-to-text engine, the audio data to converted text (Wasowicz, [0111]);
outputting for display a phonemes detail view that identifies phonemes assigned to sounds within the audio data (Wasowicz, [0111], “speak the name of each item into a microphone that is interpreted by the speech recognition software in the client computer, transmitted to the server and compared to a correct response by the speech recognition software in the server so that the scorer may determine whether or not the child correctly identified each item”; [0126]; [0013], [0127], “If the response is incorrect, the module may determine the number of consecutive errors for the particular ending sound in step 288, compare the calculated number to a predetermined number in step 290 and display a next word”);
identifying an error in the converted text by detecting a difference between the converted text and the text (Wasowicz, [0111], “speak the name of each item into a microphone that is interpreted by the speech recognition software in the client computer, transmitted to the server and compared to a correct response by the speech recognition software in the server so that the scorer may determine whether or not the child correctly identified each item”; [0126]; [0013], [0127], “If the response is incorrect, the module may determine the number of consecutive errors for the particular ending sound in step 288, compare the calculated number to a predetermined number in step 290 and display a next word”);
classifying, with a machine classifier, the error into an error category (Wasowicz, fig. 29, 740; [0056], “describe the different sounds units types, syllable types and phoneme types that may be tested using the diagnostic system since these types of sound units, syllables and phonemes are similar to the types of sound units, syllables and phonemes used in the training tools”; [0098], “indexes are then incremented as described below to analyze each incorrect response for each subtest wherein each incorrect response is compared to each error measure to determine the type of error”; [0099]);
outputting an analysis of the converted text showing a reading error made by the student at a reading-analysis interface (Wasowicz, fig. 29, 740; [0056], “describe the different sounds units types, syllable types and phoneme types that may be tested using the diagnostic system since these types of sound units, syllables and phonemes are similar to the types of sound units, syllables and phonemes used in the training tools”; [0098], “indexes are then incremented as described below to analyze each incorrect response for each subtest wherein each incorrect response is compared to each error measure to determine the type of error”; [0099]; fig. 21; pg. 25, “Phonemic”).
Wasowicz does not explicitly disclose misclassification of the error category.
Meng et al. (US 2020/0334538 A1) teaches conditional teacher-student model training. Meng further teaches receiving feedback from a user indicating that the error category is incorrect; using the feedback to generate a misclassification record; and retraining the machine classifier using data from the misclassification record as training data, wherein the training data from the misclassification record comprises the audio data of the text being read orally by the student (Meng, [0018], “incorrectly identify frames in an utterance, which can cause the performance of a student model to degrade”; [0041], “The student model 260 is trained under the supervision of the teacher model 250, wherein each model 250, 260 receives utterances in their respective domains in parallel … The parallel utterances are received by the respective teacher model 250 or student model 260, which may each correctly or incorrectly recognize the utterance as containing a given word”; [0062], “various machine learning techniques may be used to update the student model 260”; [0033], “Automatic Speech Recognition (“ASR”) task where the goal is to learn a student acoustic model”). Therefore, in view of Meng, 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 method/system, by updating the student ASR model as taught by Meng, in order to improve the student
model performance and accuracy (Meng, [0022]; [0047]).
Wasowicz does not explicitly disclose self-correction.
Alison teaches a method for analyzing running records for a reader. Alison further teaches: wherein the error category is a self-correction(Alison, pg. 5, “Cueing Systems (M-S-V)”; pg. 7, “… Appealing for help … Rerunning (going back and rereading during tricky parts … Self correcting”). Therefore, in view of Alison, 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 media/method described in Wasowicz, by providing the MSV and reading habit (i.e., self correcting) as taught by Alison, in order to monitor a reader’s word reading habits and general reading behaviors (Alison, pg. 7). A reader is self-monitoring to make sure that what he/she reads looks right, sounds right, and makes sense … self-corrects is likely doing a really good job of monitoring (Alison, pg. 10).
Wilson et al. (US 2018/0144269 A1) teaches a field of machine learning and information retrieval.
displaying, m a reading-analysis interface, a list of candidate error categories associated with the identified error; receiving a user selection of a corrected error category from among the candidate error categories;using the feedback to generate a misclassification record using the selected corrected error category as a label for the identified error (Wilson, [0072], “FIG. 12, a first classifier ("Classifier 1") can be trained using features of the training data content items based on Attribute 1 and Attribute 2. A second classifier ("Classifier 2") can be trained using features based on a different subset of the attributes. For example, the second classifier can be trained using only features based on Attribute 1”);;
in response to the user selection of the corrected error category, initiating retraining of the machine classifier using data from the misclassification record as training data to generate a student-specific machine classifier for the student, wherein the training data from the misclassification record comprises the audio data of the text being read orally by the student as input and the selected error category as the label (Wilson, [0070], “The label assigned to content items 51 and 100 are correct, while the label predicted by the classifier for content item 99”);
generate a reading competency score using the corrected error category; outputting an analysis of the converted text showing a reading error made by the student at the reading-analysis interface (Wilson, [0079], “content items 3, 5, 7 and 8 are classified incorrectly”);
outputting a second text for display at the reading interface; receiving audio data of the second text being read orally by the student; converting the audio data of the second text to converted text by the speech-to-text engine; and generating a second reading competency score for the student using the student-specific machine classifier (Wilson, [0070], “The label assigned to content items 51 and 100 are correct, while the label predicted by the classifier for content item 99 is incorrect”; col. 11, claim 7, “re-training the machine-learning classifier based on the number of correct classifications and the number of incorrect classifications”).
Therefore, in view of Wilson, 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 method described in Wasowicz, by retraining the ml as taught by Wilson, since the machine-learning classifier could be trained using both the training data and the validation data sets in the event that the classifier based only on the training data set did not achieve the requisite level of accuracy (Wilson, [0070]).
Re claims 18, 19:
18. The media of claim 17, wherein the method further comprises: outputting a classification confidence generated by the speech-to-text engine for a specific phoneme (Wasowicz, fig. 29, 740; [0056], “describe the different sounds units types, syllable types and phoneme types that may be tested using the diagnostic system since these types of sound units, syllables and phonemes are similar to the types of sound units, syllables and phonemes used in the training tools”; [0098], “indexes are then incremented as described below to analyze each incorrect response for each subtest wherein each incorrect response is compared to each error measure to determine the type of error”; [0099]).
19. The media of claim 18, wherein the method further comprises: in response to the classification confidence being less than a threshold (Wasowicz, [0133]; [0136]), adding the specific phoneme to a development list for the student; and using the development list to generate a reading-assignment recommendation for an assignment that includes the specific phoneme (Wasowicz, fig. 29, 740; [0056], “describe the different sounds units types, syllable types and phoneme types that may be tested using the diagnostic system since these types of sound units, syllables and phonemes are similar to the types of sound units, syllables and phonemes used in the training tools”; [0098], “indexes are then incremented as described below to analyze each incorrect response for each subtest wherein each incorrect response is compared to each error measure to determine the type of error”; [0099]; [0096], “The recommender 108 may use the scores and statistical information generated by the scorer, if requested by the user of the client computer, to recommend one or more training tools that may be used by the child taking the tests on the particular client computer in order to improve the child's ability in any deficient areas”).
Re claim 22:
22. The media of claim 17, wherein the training data used from the misclassification record to retrain the machine classifier further comprises a correct classification as a training label included with the audio data of the text being read orally by the student (Meng, [0018], “incorrectly identify frames in an utterance, which can cause the performance of a student model to degrade”; [0041], “The student model 260 is trained under the supervision of the teacher model 250, wherein each model 250, 260 receives utterances in their respective domains in parallel … The parallel utterances are received by the respective teacher model 250 or student model 260, which may each correctly or incorrectly recognize the utterance as containing a given word”; [0062], “various machine learning techniques may be used to update the student model 260”; [0033], “Automatic Speech Recognition (“ASR”) task where the goal is to learn a student acoustic model”).
Response to Arguments
Applicant’s arguments with respect to claim(s) 1-6, 9-19 and 21-22have been considered but are moot because the new ground of rejection does not rely on any reference applied in the prior rejection of record for any teaching or matter specifically challenged in the argument.
Applicant argues: The applicant respectfully traverses this rejection. When analyzed under the 2019 Revised Patent Subject Matter Eligibility Guidance and relevant eligibility examples, including USPTO Example 39 (Training a Neural Network for Facial Detection), the pending claims recite a specific improvement to a computer-based machine learning system that is integrated into a practical application and includes additional elements that amount to significantly more than any alleged judicial exception.
The examiner submits that machine learning (ML) is defined as a discipline of artificial intelligence (AI) that provides machines the ability to automatically learn from data and past experiences to identify patterns and make predictions with minimal human intervention. A human may classify error into an error category and retrain (refine) the machine classifier based on past data. Furthermore, a human teacher has been known to provide classifier error category from listening to pronunciation from student; error category such as: vowels, intonation, and accents. This problem is not inherently technical, nor require a lot of computational resources. The human has been known to adjust reading level based on the reading error and competency score.
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 JACK YIP whose telephone number is (571)270-5048. The examiner can normally be reached Monday thru Friday; 9:00 AM - 5:00 PM EST.
Examiner interviews are available via telephone, in-person, and video conferencing using a USPTO supplied web-based collaboration tool. To schedule an interview, applicant is encouraged to use the USPTO Automated Interview Request (AIR) at http://www.uspto.gov/interviewpractice.
If attempts to reach the examiner by telephone are unsuccessful, the examiner’s supervisor, XUAN THAI can be reached at (571) 272-7147. 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.
/JACK YIP/Primary Examiner, Art Unit 3715