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 .
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.
The Manual of Patent Examining Procedure (MPEP) provides detailed rules for determining subject matter eligibility for claims in §2106. Those rules provide a basis for the analysis and finding of ineligibility that follows. MPEP §2106(III) states that examiners should determine whether a claim satisfies the criteria for subject matter eligibility by evaluating the claim in accordance with the flowchart in this section.
Claims 1-20 are rejected under 35 U.S.C. 101. The claimed invention is directed to unpatentable subject matter because the claimed invention recites a judicial exception (i.e., a law of nature, a natural phenomenon, or an abstract idea) without significantly more. The examiner analyzes the instant claims according to a flowchart for subject matter eligibility test for products and processes (MPEP 2106).
Eligibility Step 1 (MPEP 2106.03, Statutory category):
Claims 1-20 are directed to a method. The claims 1-20 fall into one of the four statutory categories of invention (YES branch of step 1).
Eligibility Step 2A, Prong One (does a claim recites a judicial exception?) (MPEP 2106.04(a) – (c)):
Step 2A is a two-prong inquiry, in which examiners determine in Prong One whether a claim recites a judicial exception, and if so, then determine in Prong Two if the recited judicial exception is integrated into a practical application of that exception. Together, these prongs represent the first part of the Alice/Mayo test, which determines whether a claim is directed to a judicial exception (See a flowchart in MPEP 2106.04(II)(A)). In the prone one of the two prong inquiry, the above limitations recited in claims are directed to at least one of groups of abstract ideas (MPEP 2106.04(a), “Mathematical concepts”, “Certain methods of organizing human activity”, “Mental Processes”). It should be noted that these groupings are not mutually exclusive, i.e., some claims recite limitations that fall within more than one grouping or sub-grouping (MPEP 2106.04(a)(2)).
Although claims 1-20 fall into one of the four statutory categories the patent eligible subject matter, independent claims recite a number of steps of (“receiving audio data…”, “processing audio data”, “determining …”, “causing …”). These limitations fall into a judicial exception (MPEP 2106.04 (II), “laws of nature”, “natural phenomena” and “abstract idea”). The Supreme Court has explained that the judicial exceptions reflect the Court’s view that abstract ideas, laws of nature, and natural phenomena are "the basic tools of scientific and technological work", and are thus excluded from patentability because "monopolization of those tools through the grant of a patent might tend to impede innovation more than it would tend to promote it." Alice Corp., 573 U.S. at 216, 110 USPQ2d at 1980. It should be noted that there are no bright lines between the types of exceptions, and that many of the concepts identified by the courts as exceptions can fall under several exceptions (MPEP 2106.04 (I) and (II)).
In light of the disclosure (Spec. [0004], [0015-0016], Fig. 1B, Fig. 3C), the claimed inventions are related to recognizing a user’s a voice command by a virtual assistant (i.e., [0001], a chatbot). In an illustration, the virtual assistant misrecognized a user’s voice command “play Madonna on app A” as “play My Donna on app A”. The user corrects the recognition error.
Although the disclosure describes a virtual assistant to recognize a user voice request and perform a function of the user’s voice request, the recited claim limitations could be reasonably interpreted as a person (e.g., a manager) tells another person (e.g., a secretary) to do certain work. For example, a method of claim 1 could be reasonably interpreted as:
receiving, via a client device, audio data capturing a spoken request of a user (a secretary receives a phone call from a manager);
processing the audio data capturing the spoken request to generate a speech recognition of the spoken request (the secretary recognizes manager’s voice request: “send an email to Cathy”);
determining whether the speech recognition of the spoken request includes any mis-transcribed phrase (the secretary suspects that she might misrecognize a recipient’s name: “Cathy” or “Kathy”);
in response to determining that the user request includes a potentially mis-transcribed phrase (in response to a possible misrecognition for recipient’s name of “Kathy” as “Cathy”),
determining one or more candidate phrases that are phonetically similar to the potentially mis-transcribed phrase (the secretary determines that “Kathy” and “Cathy” has similar sounding, the secretary thought that the manager may want to send an email to “Kathy”, not to “Cathy”), and
monitoring for additional audio data capturing any of the one or more candidate phrases that are phonetically similar to the potentially mis-transcribed phrase (the secretary hears her manager is talking about “Kathy” starting with “K” not “C”); and
in response to receiving a user input that includes or selects a particular candidate phrase, of the one or more candidate phrases (in response to the manager indicates the name is “Kathy”, not “Cathy”),
causing an action that corresponds to the particular candidate phrase, but not corresponding to the potentially mis-transcribed phrase, to be performed (the secretary sends an email to Kathy but not sending an email to “Cathy”).
If claim 1 is patented, the secretary would infringe the patent if she is doing her routine work. Independent claim 7 and 15 include some more details such as “performing natural language understanding”, “enter phonetically restricted mode”. These limitations are also mental process. For example, a human can perform natural language processing in his / her mind to determine if a word makes sense (e.g., a generated transcript “cook a flight” does not make any sense. It must mis-transcribed a word “book” to “cook”). The human can also try to distinguish words of similar sounding such as listens to spelling, which is a claimed “enter phonetically restricted mode”.
Claim 2 recites limitation related to a user speaks a name, which is a mental process. Claim 3 is directed to writing down two similar sounding words and asking a selection, these limitations could also be interpreted as a mental process with a piece of paper and pen. Claim 4 is asking a user selection, which is a mental process. Claim 5 is related to obtaining an acoustic score to determine if a word is misrecognized, the claimed feature is also a mental process based on similar sounding. Claim 6 is related to using natural language understanding to determine if a word is misrecognized. A human can determine based on language knowledge if a recognized word make sense. Dependent claims 8-14 and 16-20, although depend from claim 7 or claim 15, respectively, recite limitations to determine if a word is misrecognized and asks a user to select a correct word. All these features could be performed by human mind with a piece of paper / a pen.
The courts consider a mental process (thinking) that “can be performed in the human mind, or by a human using a pen and paper” to be an abstract idea. CyberSource Corp. v. Retail Decisions, Inc., 654F.3d 1366, 1372, 99 USPQ2d 1690, 1695 (Fed. Cir.2011). If a claim recites a limitation that can practically be performed in the human mind, with or without the use of a physical aid such as pen and paper, the limitation falls within the mental processes grouping, and the claim recites an abstract idea. See, e.g., Benson, 409 U.S. at 67, 65, 175 USPQ at 674-75,674. If the claimed invention is described as a concept that is performed in the human mind and applicant is merely claiming that concept performed 1) on a generic computer, or 2) in a computer environment, or 3) is merely using a computer as a tool to perform the concept.
As analyzed above, the recited limitations are directed to determining if a word is misrecognized due to similar sounding words (e.g. “Kathy” vs “Cathy”; “My Donna” vs. “Madonna”). If a word is misrecognized, asking a user to select a correct word from several candidate words. These claim limitations are mental process.
In these situations, the claim is considered to recite a mental process. The Court concluded that the algorithm could be performed purely mentally even though the claimed procedures “can be carried out in existing computers long in use, no new machinery being necessary.” The claims therefore recited an abstract idea, despite the fact that the claimed steps were performed on a computer. 887 F.3d at 1385, 126 USPQ2d at 1504.
Eligibility Step 2A, Prong two (integrated into a practical application? MPEP 2106.04(d)).
Since the claimed invention falls into a judicial exception according above analysis (YES branch of PRONG ONE in the step 2A), a claim that is directed to a judicial exception must be evaluated to determine whether the claim recite additional elements that integrate the judicial exception into a practical application (MPEP 2106.04(II)(A)(2)). Prong Two asks whether the claim recite additional elements that integrate the judicial exception into a practical application. In Prong Two, examiners evaluate whether the claim as a whole integrates the exception into a practical application of that exception. Court in Gottschalk v. Benson ‘‘held that simply implementing a mathematical principle on a physical machine, namely a computer was not a patentable application of that principle. Accordingly, after determining that a claim recites a judicial exception in Step 2A Prong One examiners should evaluate whether the claim as a whole integrates the recited judicial exception into a practical application of the exception in Step 2A Prong Two. For a claim reciting a judicial exception to be eligible, the additional elements (if any) in the claim must "transform the nature of the claim" into a patent-eligible application of the judicial exception, Alice Corp., 573 U.S. at 217, 110 USPQ2d at 1981, either at Prong Two or in Step 2B. If there are no additional elements in the claim, then it cannot be eligible.
Eligibility Step 2B (Inventive concept / significantly more consideration; MPEP 2106.05).
MPEP §2106.05 describes step 2B test to determine whether a claim amounts to significantly more. The second part of the Alice/Mayo test is often referred to as a search for an inventive concept. Alice Corp. Pty. Ltd. v. CLS Bank Int'l, 573 U.S. 208, 217, 110 USPQ2d 1976, 1981 (2014). The Supreme Court has identified a number of considerations as relevant to the evaluation of whether the claimed additional elements amount to an inventive concept (See MPEP §2106.05(I)(A)). It is notable that mere physicality or tangibility of an additional element or elements is not a relevant consideration in Step 2B. As the Supreme Court explained in Alice Corp., mere physical or tangible implementation of an exception is not in itself an inventive concept and does not guarantee eligibility.
The Supreme Court has identified a number of considerations as relevant to the evaluation of whether the claimed additional elements amount to an inventive concept. By considering limitations recited in the instant claims, the claims do not improve the functions of a computer, or any other technology or technical field. The claims also do not apply the judicial exception with, or by use of, a particular machine. The claims also do not have effecting a transformation or reduction of a particular article to a different state or thing. The claims fail to include a specific limitation other than what is well-understood, routine, conventional activity in the field, or adding unconventional steps that confine the claim to a particular useful application. The recited “processor” / “memory” are well-understood, routine and conventional in the field. Therefore, that recited element does not amount to significantly more than an abstract idea.
Please notes simply appending well-understood, routine, conventional activities previously known to the industry, specified at a high level of generality, to the judicial exception, e.g., a claim to an abstract idea requiring no more than a generic computer to perform generic computer functions that are well-understood, routine and conventional activities previously known to the industry, as discussed in Alice Corp., 573 U.S. at 225, 110 USPQ2d at 1984. The court also found “adding insignificant extra-solution activity to the judicial exception” or “generally linking the use of the judicial exception to a particular technological environment or field of use” is not enough to be qualify as “significantly more” considerations.
By reviewing limitations recited in the claims, none of the limitations meet the significantly more considerations. Therefore, claims are directed to unpatentable subject matter and are rejected under 35 U.S.C. 101 (MPEP §2106, flowchart, Step 2B, NO branch).
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 1-20 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 pre-AIA the applicant regards as the invention.
Claim 1 recites “the user request includes…”. Which has insufficient antecedent basis.
An antecedent limitation recites “a spoken request of a user” or “the spoken request”, a claimed “the user request” could refer to a user typed request, not necessarily a previously mentioned “a spoken request”.
By reviewing the disclosure, it appears applicant intended to express a meaning of: “the speech recognition of the spoken request includes potentially mis-transcribed phrase”, rather than a claimed “the user request includes a potentially mis-transcribed phrase”. Claims 2-5 include limitations of claim 1. Claims 2-5 fail to remedy deficiency of claim 1.
Claims 7 and 15 recites “causing the automated assistant to…”, which has insufficient antecedent basis. The antecedent limitations never mention any automated assistant. Dependent claims 8-14 and 16-20 include limitations of their corresponding independent claims 7 and 15, respectively.
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 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 of this title, 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 1-5 are rejected under 35 U.S.C. §103 as being unpatentable over Lewis et al. (US PG Pub. 2021/0074277, referred to as Lewis) in view of Waibel et al. (US Pat. 5,855,000, referred to as Waibel).
Lewis discloses correcting the mis-recognized text by an automatic speech recognition system. The recognized text may have misrecognized words due to similar pronunciations (Fig. 1, when a user issues a voice command: “Remember to book flight for July 10”, the system misrecognized the voice command as “Remember to cook flight for July 10”). The system uses language models, sound-similarity models, character matching, eye-gaze tracking, and even stylus pointing to locate a likely recognition error. The user then enters the corrected text, and the system places it in the right spot. The system may also highlight suspicious parts of the transcription so the user can review them faster. Fig. 3A (replicated below) shows a voice command is mis-transcribed due to similar pronunciation.
PNG
media_image1.png
570
1182
media_image1.png
Greyscale
Waibel discloses a speech recognition system (Waibel, Fig. 1). Waibel discloses that if a user notices a speech recognition error, the user could correct the recognition error using many techniques. For example, if the speech recognition system misrecognized “Monday” as “One day”, the user could correct the recognition error by repeating the word (Waibel, Fig. 4), spelling out the word (Waibel, Fig. 5, replicated below) or correcting the error using handwriting (Waibel, Fig. 6).
PNG
media_image2.png
610
600
media_image2.png
Greyscale
Regarding claim 1, Lewis discloses a method implemented by one or more processors (Lewis, [0010], Fig. 1, a computer implemented voice command processing system to correct speech recognition errors when processing a user issued a voice command), the method comprising:
receiving, via a client device, audio data capturing a spoken request of a user (Lewis, [0010], Fig. 1, #12);
processing the audio data capturing the spoken request to generate a speech recognition of the spoken request (Lewis, [0010], Fig. 1, #12, processing a user’s voice command, when generating transcript, misrecognized “book flight” as “cook flight” due to similar sounding);
determining whether the speech recognition of the spoken request includes any mis-transcribed phrase (Lewis, [0021], determining a portion of the transcript is likely contain an error due to similar pronunciations);
in response to determining that the user request includes a potentially mis-transcribed phrase (Lewis, [0016], [0019-0020], allow a user to provide a correction),
determining one or more candidate phrases that are phonetically similar to the potentially mis-transcribed phrase (Lewis, [0021], [0026-0027], based on confidence levels, some similar words maybe mis-transcribed due similar sounding words, e.g., cook [Wingdings font/0xE8] book or Poland [Wingdings font/0xE8] Portland), and
monitoring for additional Lewis, [0016], [0019], [0036], Fig. 3A/3B, user provides correction to the mis-transcribed words, e.g., a user type “IC” to correct a mis-transcribed “I see”); and
in response to receiving a user input that includes or selects a particular candidate phrase, of the one or more candidate phrases (Lewis, Lewis, [0016], [0019], [0036], Fig. 3A/3B, user provides correction to the mis-transcribed words; Fig. 1, #127a),
causing an action that corresponds to the particular candidate phrase, but not corresponding to the potentially mis-transcribed phrase, to be performed (Lewis, [0021], [0031], [0036], performing correct function of a voice command “book flight” instead of misrecognized “cook flight”).
Lewis discloses when processing a user’s voice command, if confidence level is low due to similar sounding words, the system highlights a possible misrecognized word due to similar sounding (Lewis, [0033], [0037]). The user could provide a correction or selection (Lewis, [0051], [0054]). Lewis does not explicitly disclose a user issues additional voice to correct error and fails to discloses “monitoring for additional audio data”.
Waibel discloses the when a recognition occurs, a user could speak the misrecognized word again (Waibel, Col. 6, lines 24-35, a user respeak a misrecognized word; Col. 8, lines 14-36, Fig. 4), spell out the misrecognized word (Waibel, Col. 8, lines 14-38, Fig. 5).
Both Lewis and Waibel are related to recognizing speech and correcting speech recognition error. It would have been obvious to a person having ordinary skill in the art at the time the invention was filed to modify Lewis’s teaching with Waibel’s teaching to monitor user’s additional voice input for correcting the recognition error. One having ordinary skill in the art would have been motivated to make such a modification to improve recognition accuracy (Waibel, Col. 15, table 2, table 3, shows speech recognition improvement).
Regarding claim 2, Lewis in view of Waibel further discloses the user input is an additional spoken request that includes the particular candidate phrase (Waibel, Col. 10, lines 5-25, Fig. 7, Fig. 9, a user repeats the misrecognized word).
Regarding claim 3, Lewis in view of Waibel further discloses:
in response to determining that the user request includes a potentially mis-transcribed phrase (Lewis, [0043-0044], Fig. 3A, #304, Fig. 3B, #404),
generating one or more selectable elements each displaying a respective candidate phrase from the one or more candidate phrases (Lewis, [0043-0044], Fig. 3A, #304, Fig. 3B, #404; Waibel, Col. 2, lines 10-15, Col. 6, lines 35-45, selecting from a n-best list), and
causing the one or more selectable elements to be visually displayed to the user (Lewis, [0043-0044], Fig. 3A, #304, Fig. 3B, #404; Waibel, Col. 2, lines 10-15, Col. 6, lines 35-45, selecting from a n-best list).
Regarding claim 4, Lewis in view of Waibel further discloses:
the user input is a user selection of a selectable element, of the one or more displayed selectable elements, that displays the particular candidate phrase (Lewis, [0043-0044], Fig. 3A, #304, Fig. 3B, #404; Waibel, Col. 2, lines 10-15, Col. 6, lines 35-45, selecting from a n-best list).
Regarding claim 5, Lewis in view of Waibel further discloses:
generating a speech recognition score for each word in the speech recognition of the spoken request (Lewis, [0033], Waibel, Fig. 9), and
determining whether the speech recognition of the spoken request includes any mis-transcribed phrase based on the speech recognition score for each word in the speech recognition of the spoken request (Lewis, [0032-0033], comparing score with a threshold to determining if a word is misrecognized).
Claims 6-20 are rejected under 35 U.S.C. §103 as being unpatentable over Lewis in view of Waibel, and further in view of Printz (US PG Pub. 2019/0147876, referred to as Printz).
Independent claims 7 and 15 are narrower than a broad independent claim 1 by including additional limitations. For limitations that are the same limitations recited in independent claim 1, please refer to explanations for claim 1. In the following section, the additional limitations that are not recited in claim 1 are addressed.
Claims 7 and 15 recites “in response to …, the automated assistant to enter phonetically restricted listening state”.
Waibel discloses when a recognition error is detected, the speech recognition system enters a spelling recognition mode (a claimed “a phonetically restricted listening mode”) to recognize spelling correction for the misrecognized word (Waibel, Col. 5, lines 36-50, Fig. 5, when “Monday” is misrecognized as “One day”, spelling out Monday).
Independent claims 7 and 15 further recites “performing natural language understanding (NLU) of the speech recognition …”, “determining, based on performing the natural language understanding (NLU) of the speech recognition, whether the speech recognition of the spoken user request includes any mis-transcribed phrase”
Lewis discloses using natural language processing (NLP) techniques (Lewis, [0064]). Lewis does not discloses determining a speech recognition error using NLU technique.
Printz discloses recognizing a voice request using both automatic speech recognition (ASR) and natural language understanding (NLU). Printz discloses a first speech recognizer makes a rough transcript of the whole utterance, and a language-understanding module decides whether part of that utterance likely contains a named entity. If so, the system isolates that portion or otherwise marks it as a target span. A second recognizer then uses a smaller, specialized vocabulary or grammar tailored to the likely entity type. The second recognizer can be adapted very quickly, so it can recognize names that were not in the original large vocabulary. Printz discloses
performing natural language understanding (NLU) of the speech recognition to determine a first action or first content responsive to the spoken user request (Printz, [0130], [0147-0148], [0187], Fig. 8, #855, using NLU to detect a word in the recognized sentence is a business name);
determining, based on performing the natural language understanding (NLU) of the speech recognition, whether the speech recognition of the spoken user request includes any mis-transcribed phrase (Printz, [0147-0148], Fig. 8, #855, using NLU to detect a word in the recognized sentence is business name, but the initial transcription may be incorrect for name entities; Fig. 16 shows when recognizing “Show me the details for L’Avenida”, the NLU determines “L’Avenida” is a business name);
Lewis, Waibel and Printz are related to recognizing speech and correcting speech recognition error. It would have been obvious to a person having ordinary skill in the art at the time the invention was filed to modify Lewis’s teaching with Waibel’s teaching to use a phonetically restricted listening state to recognize spelling for a word. It would also have been obvious to a person having ordinary skill in the art at the time the invention was filed to modify Lewis in view of Waibel’s teaching with Printz’s teaching to use NLU to determine a misrecognized word is a name entity and employs a secondary recognizer specially for recognizing names. One having ordinary skill in the art would have been motivated to make such a modification to improve recognition accuracy (Waibel, Col. 15, table 2, table 3, shows speech recognition improvement; Printz, [0105], [0108], obtain higher accuracy of speech recognition).
Regarding claim 6, which depend from claim 1, but include additional limitations added to independent claim 7 (performing natural language understanding (NLU) on the speech recognition of the spoken request to generate a NLU score, and determining whether the speech recognition of the spoken request includes any mis-transcribed phrase based on the NLU score).
The added limitations are taught in Printz reference. Claim 6 is rejected based on the same rational as explained above for claim 7.
Regarding claim 8, Lewis in view of Waibel and Printz further discloses:
receiving, via the client device, additional audio data capturing a spoken utterance (Waibel, Fig. 5, spelling out a misrecognized word);
processing the additional audio data capturing the spoken utterance to determine that the spoken utterance includes a particular candidate phrase, of the one or more candidate phrases, that is phonetically similar to the potentially mis-transcribed phrase (Lewis, Fig. 3A, correcting a mis-transcribed location name “Poland” to “Portland”’ Waibel, Fig. 5, correcting mis-transcribed word “Monday” by spelling out the word); and
in response to determining that spoken utterance includes the particular candidate phrase that is phonetically similar to the potentially mis-transcribed phrase (Lewis, Fig. 3A, Waibel, Fig. 4-6),
determining, based on the speech recognition of the spoken user request and the particular candidate phrase, a second action that is different from the first action, or second content that is different from the first content, as being responsive to the spoken user request (Lewis, Fig. 1, booking a flight, not cooking a flight; Fig. 3A, booking a flight to Portland, not to Poland), and
causing the second action to be performed, or the second content to be displayed (Lewis, Fig. 1 and Fig. 3A).
Regarding claims 9 and 16, Lewis in view of Waibel and Printz further discloses:
replacing, in the speech recognition of the spoken user request, the potentially mis-transcribed phrase with the particular candidate phrase, to generate a user query (Lewis, Fig. 1, replacing misrecognized word “cook” with “book”, and book a flight), and
determining the second action or the second content based on the generated user query (Lewis, Fig. 1, Fig. 3A, book a flight to Portland, not Poland).
Regarding claims 10 and 17, Lewis in view of Waibel and Printz further discloses:
causing the first action to be performed, or the first content to be displayed (Lewis, [0016], Fig. 1),
generating a message asking for confirmation of the spoken user request, or a messaging asking for confirmation of the first action or the first content (Lewis, [0021], [0033]), and
causing the message to be rendered via the client device prior to receiving the additional audio data capturing the spoken utterance (Lewis, [0019], a system identifying a portion that should be correct, displaying a correction box).
Regarding claim 11, Lewis in view of Waibel and Printz further discloses:
pausing or terminating the first action prior to causing the second action to be performed, OR
causing the first content to disappear prior to causing the second content to be displayed (Lewis, Fig. 3B, #405, replacing the misrecognized word; Waibel, Col. 11, lines 5-10, replace the highlighted section with new top-choice).
Regarding claims 12 and 18, Lewis in view of Waibel and Printz further discloses:
performing natural language understanding (NLU) of the speech recognition comprises: performing natural language understanding of the speech recognition to generate one or more NLU scores for the speech recognition (Printz, [0165-0167], NLU confidence scor;), and
determining whether the speech recognition of the spoken user request includes any mis-transcribed phrase is based on the one or more NLU scores (Printz, [0147-0148], [0165-0167]).
Regarding claims 13 and 19, Lewis in view of Waibel and Printz further discloses:
performing speech recognition of the audio data capturing the spoken user request to generate one or more speech recognition scores for the speech recognition (Lewis, [0032-0033]. Waibel, Fig 9), and
determining whether the speech recognition of the spoken user request includes any mis-transcribed phrase is further based on the one or more speech recognition scores (Lewis, [0032-0033]. Waibel, Fig 9).
Regarding claims 14 and 20, Lewis in view of Waibel and Printz further discloses:
determining the one or more candidate phrases that are phonetically similar to the potentially mis-transcribed phrase based on a sequence of phonemes of the potentially mis-transcribed phrase (Lewis, Fig. 1, “cook” and “book” are phonetically similar, Fig. 3B, “IC” vs. “I see”).
Conclusion
The prior art made of record and not relied upon is considered pertinent to applicant's disclosure. The examiner discovered several relevant prior art references that are related to one or more concepts disclosed by the instant application. These references are included in the attached PTO-892 form for completeness of the record.
Any inquiry concerning this communication or earlier communications from the examiner should be directed to Jialong He, whose telephone number is (571) 270-5359. The examiner can normally be reached on Monday – Friday, 8:00AM – 4:30PM, EST.
If attempts to reach the examiner by telephone are unsuccessful, the examiner's supervisor, Pierre Desir can be reached on (571) 272-7799. The fax phone number for the organization where this application or proceeding is assigned is 571-273-8300.
Information regarding the status of an application may be obtained from the Patent Application Information Retrieval (PAIR) system. Status information for published applications may be obtained from either Private PAIR or Public PAIR. Status information for unpublished applications is available through Private PAIR only. For more information about the PAIR system, see http://pair-direct.uspto.gov. Should you have questions on access to the Private PAIR system, contact the Electronic Business Center (EBC) at 866-217-9197 (toll-free). If you would like assistance from a USPTO Customer Service Representative or access to the automated information system, call 800-786-9199 (IN USA OR CANADA) or 571-272-1000.
/JIALONG HE/Primary Examiner, Art Unit 2659
Prosecution Insights