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 .
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 3, 14 and 16-20 are 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.
Regarding claims 3 and 14, the phrase "or other graphic features" renders the claim(s) indefinite because the claim(s) include(s) elements not actually disclosed (those encompassed by "other graphic features"), thereby rendering the scope of the claim(s) unascertainable. See MPEP § 2173.05(d).
Claim 16 recites the limitation "an updated phoneme sequence" in line 12 of the claim. Since this recitation is the first recitation of “phoneme sequence” in the claim, it is unclear what “phoneme sequence” is being “updated”. That is, the claim does not indicate whether the recited “phoneme sequence” is limited to phonemes associated with the substitute phrase, or if the entire plurality of tokens and phrases representing an utterance are associated with a “phoneme sequence”. It is also possible that the recited “phoneme sequence” is independent of the recited tokens and phrases. Since it cannot be determined how the recited “phoneme sequence” relates to the other limitations of the claim, claim 16 is considered indefinite.
Claims 17-20 are considered indefinite for their dependence on claim 16.
Additionally, claim 17 recites the limitation "the substitute token" in line 2 of the claim. There is insufficient antecedent basis for this limitation in the claim.
Claim 18 recites the limitation "the query sentence" in line 2 of the claim. There is insufficient antecedent basis for this limitation in the claim.
Claim 20 recites the limitation "the respective sentence-level acoustic scores" in line 3 of the claim. There is insufficient antecedent basis for this limitation in the claim. Additionally, while lines 3-4 of the claim recite “individual phoneme sequences”, there is insufficient antecedent basis for “the plurality of phoneme sequences” recited in line 5 of the claim.
Claim Rejections - 35 USC § 102
The following is a quotation of the appropriate paragraphs of 35 U.S.C. 102 that form the basis for the rejections under this section made in this Office action:
A person shall be entitled to a patent unless –
(a)(1) the claimed invention was patented, described in a printed publication, or in public use, on sale, or otherwise available to the public before the effective filing date of the claimed invention.
Claim(s) 1-5 is/are rejected under 35 U.S.C. 102(a)(1) as being anticipated by Milstein (U.S. Patent Application Pub. No. 2012/0022865).
In regard to claim 1, Milstein discloses a computer-implemented method for speech recognition, comprising:
generating, at an acoustic model of an automatic speech recognition system, a phoneme sequence of a received utterance (speech is input to an automated speech recognition system, which assigns a sequence of phonemes to short segments of the speech, paragraph [0002]);
generating a token sequence that represents the phoneme sequence based on a pronunciation dictionary (the phonemes are then compared to a grammar of known words to generate a sequence of words, phrases, or sentence tokens, paragraphs [0002] and [0023]);
assigning respective token confidence scores to individual tokens in the token sequence, wherein a token confidence score represents a level of confidence of the correct representation of a tokenized word (confidence scores are assigned to each token, paragraph [0023]); and
indicating, on a graphic user interface, the respective token confidence scores associated with the individual tokens in the token sequence (see Fig. 8, confidence scores, represented as a percentage, are displayed for each transcribed token, paragraphs [0049-0050]).
In regard to claim 2, Milstein discloses the respective token confidence scores are used to prioritize transcription for labeling (tokens with low confidence values are provided to a human transcriptionist for labeling with the correct token, paragraph [0023]).
In regard to claim 3, Milstein discloses displaying the respective token confidence scores in various colors, fonts, or other graphic features (low confidence scores are displayed through different font sizes and styles, paragraph [0049]).
In regard to claim 4, Milstein discloses the respective token confidence scores are used to detect labeling errors (the confidence scores identify questionable utterances having a likelihood of being associated with an incorrect transcription, paragraph [0049]).
In regard to claim 5, Milstein discloses determining that the pronunciation dictionary does not comprise a new token with a low token confidence score (a questionable utterance with a low confidence score is manually transcribed, paragraphs [0057-0058]); and
adding the new token to the pronunciation dictionary (once a new transcription has been assigned a threshold number of times, the new transcribed value is added to the grammar, paragraph [0059]).
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.
Claim(s) 6 and 9-18 is/are rejected under 35 U.S.C. 103 as being unpatentable over Milstein, in view of Chen (U.S. Patent No. 8,812,316).
In regard to claim 6, Milstein discloses a computer-implemented method for speech recognition, comprising:
receiving, at an automatic speech recognition system, an utterance (speech is input to an automated speech recognition system, paragraph [0002]);
generating, based on an acoustic model, a phoneme sequence of the utterance (an acoustic model of the speech recognition system assigns a sequence of phonemes to short segments of the speech, paragraph [0002]);
assigning respective token confidence scores to individual tokens in the phoneme sequence, wherein a token confidence score represents a level of confidence of the correct representation of a tokenized word (confidence scores are assigned to each token, paragraph [0023]);
determining that a first confidence score associated with a token is lower than a predetermined threshold (tokens with a confidence below a threshold are determined to be questionable utterances, paragraph [0031]);
indicating, on a graphic user interface, the respective token confidence scores associated with the individual tokens in the token sequence (see Fig. 8, confidence scores, represented as a percentage, are displayed for each transcribed token, paragraphs [0049-0050]).
Milstein further discloses embodiments wherein additional grammars are utilized to determine tokens (paragraph [0060]), but do not expressly disclose substituting tokens from the additional grammars.
Chen discloses a method for speech recognition, comprising assigning first respective token confidence scores to tokens, then determining a substitute token associated with a second confidence score that is higher than the first confidence score (a plurality of interpreters determines confidence scores associated with replacing initial tokens with substitute tokens, column 8, line 39 to column 9, line 5); and
updating the token sequence by replacing the token with the substitute token (the highest confidence tokens from each interpreter are merged in a non-overlapping manner to repair the originally transcribed tokens, column 9, lines 5-15).
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to automatically determine a substitute token associated with a second confidence score higher than the first confidence score and update the token sequence by replacing the token with the substitute token, because it would allow the text to be repaired automatically without requiring a user’s input, as taught by Chen (column 2, lines 28-41).
In regard to claim 9, Milstein discloses a confidence score model assigns the respective token confidence scores to individual tokens in the token sequence (confidence scores are assigned to each token, paragraph [0023]).
In regard to claim 10, Milstein does not disclose a translation model.
Chen discloses a translation model determines the substitute token and update the token sequence by replacing the token with the substitute token (a plurality of interpreters determines confidence scores associated with replacing initial tokens with substitute tokens, column 8, line 39 to column 9, line 5).
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to use a translation model to determine the substitute token and update the token sequence by replacing the token with the substitute token, because it would allow the text to be repaired automatically without requiring a user’s input, as taught by Chen (column 2, lines 28-41).
In regard to claim 11, Milstein discloses the token confidence scores are based on one or more of a token sequence probability analysis, acoustic probability analysis, and semantic analysis (confidence scores are assigned to each utterance based on automatic speech recognition using, for example, token sequence models such as n-grams, paragraph [0031]).
In regard to claim 12, Milstein discloses generating, based on the phoneme sequence, a text transcription of the utterance (a text representation of the corresponding voice utterance, paragraph [0031]).
In regard to claim 13, Milstein discloses the respective token confidence scores are used to prioritize transcription for labeling (tokens with low confidence values are provided to a human transcriptionist for labeling with the correct token, paragraph [0023]).
In regard to claim 14, Milstein discloses displaying the respective token confidence scores in various colors, fonts, or other graphic features (low confidence scores are displayed through different font sizes and styles, paragraph [0049]).
In regard to claim 15, Milstein discloses the respective token confidence scores are used to detect labeling errors (the confidence scores identify questionable utterances having a likelihood of being associated with an incorrect transcription, paragraph [0049]).
In regard to claim 16, Milstein discloses a computer-implemented method for speech recognition, comprising:
receiving, at an acoustic model of an automatic speech recognition system, plurality of tokens and phrases representing an utterance (speech is input to an automated speech recognition system, paragraph [0002]; comprising a plurality of utterances and phrases, paragraph [0031]; and an acoustic model of the speech recognition system assigns a sequence of phonemes to short segments of the speech, paragraph [0002]);
assigning, by a confidence score model, respective token confidence scores and phrase confidence scores to the plurality of tokens and phrases, wherein a token confidence score and a phrase confidence score represents a level of confidence of the correct representation of a word or phrase (confidence scores are assigned to each token and phrase using word models, phrase models, etc., paragraphs [0023] and [0031]);
determining that a first confidence score associated with a phrase is lower than a predetermined threshold (tokens with a confidence below a threshold are determined to be questionable utterances, paragraph [0031]);
indicating, on a graphic user interface, the respective token confidence scores associated with the individual tokens in the token sequence (see Fig. 8, confidence scores, represented as a percentage, are displayed for each transcribed token, paragraphs [0049-0050]).
Milstein further discloses embodiments wherein additional grammars are utilized to determine tokens (paragraph [0060]), but do not expressly disclose substituting tokens from the additional grammars.
Chen discloses a method for speech recognition, comprising assigning first respective token confidence scores to tokens, then determining, by a translation model, a substitute phrase associated with a second confidence score that is higher than the first confidence score (a plurality of interpreters determines confidence scores associated with replacing initial tokens with substitute tokens, column 8, line 39 to column 9, line 5); and
replacing the phrase with the substitute phrase to generate an updated phoneme sequence (the highest confidence tokens from each interpreter are merged in a non-overlapping manner to repair the originally transcribed tokens, column 9, lines 5-15).
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to automatically determine a substitute token associated with a second confidence score higher than the first confidence score and update the token sequence by replacing the token with the substitute token, because it would allow the text to be repaired automatically without requiring a user’s input, as taught by Chen (column 2, lines 28-41).
In regard to claim 17, Milstein does not disclose a translation model.
Chen discloses a translation model determines the substitute token and update the token sequence by replacing the token with the substitute token (a plurality of interpreters determines confidence scores associated with replacing initial tokens with substitute tokens, column 8, line 39 to column 9, line 5).
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to use a translation model to determine the substitute token and update the token sequence by replacing the token with the substitute token, because it would allow the text to be repaired automatically without requiring a user’s input, as taught by Chen (column 2, lines 28-41).
In regard to claim 18, Milstein discloses receiving the utterance of the query sentence (a voice stream, paragraph [0028]);
generating a phoneme sequence of the utterance (the automated speech recognition system assigns phonemes to short segments of the input speech, paragraph [0002]); and
segmenting the phoneme sequence into a plurality of tokens and phrases based on a pronunciation dictionary, wherein a phrase comprises one or more tokens (the phonemes are compared to a grammar of known words, phrases and sentences to determine tokens, paragraphs [0002] and [0031]).
Allowable Subject Matter
Claims 7-8 are objected to as being dependent upon a rejected base claim, but would be allowable if rewritten in independent form including all of the limitations of the base claim and any intervening claims.
The following is a statement of reasons for the indication of allowable subject matter:
Milstein, Chen, and the additional prior art of record do not disclose or suggest assigning respective sentence-level acoustic scores to individual phoneme sequences to indicate the respective likelihood of correctness to represent the utterance, as required by claim 7.
Claim 8 would be allowable for its dependence on claim 7.
Claims 19 would also be allowable if rewritten in independent form including all of the limitations of the base claim and any intervening claims, and if the 35 U.S.C. 112 rejections were overcome.
Claim 20 would also be allowable if rewritten in independent form including all of the limitations of the base claim and any intervening claims, and if the 35 U.S.C. 112 rejections were overcome, and if amended to depend on claim 19, rather than claim 16.
Conclusion
The prior art made of record and not relied upon is considered pertinent to applicant's disclosure.
Wang et al. disclose an additional method that determines substitute tokens and phrases based on confidence scores. Ben-David et al., Davis et al., and Kobal et al. disclose additional methods that display indications of token confidence.
Any inquiry concerning this communication or earlier communications from the examiner should be directed to BRIAN LOUIS ALBERTALLI whose telephone number is (571)272-7616. The examiner can normally be reached M-F 8AM-3PM, 4PM-5PM.
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.
BLA 10/1/26
/BRIAN L ALBERTALLI/ Primary Examiner, Art Unit 2656
Prosecution Insights