DETAILED ACTION
Introduction
This office action is in response to Applicant’s submission filed on April 28, 2025.
Claims 1-20 are pending in the application. As such, claims 1-20 have been examined.
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 .
Drawings
The drawings were received on April 28, 2025, and modified on February 9, 2026. These drawings have been accepted and considered by the Examiner.
Claim Interpretation
The following is a quotation of 35 U.S.C. 112(f):
(f) Element in Claim for a Combination. – An element in a claim for a combination may be expressed as a means or step for performing a specified function without the recital of structure, material, or acts in support thereof, and such claim shall be construed to cover the corresponding structure, material, or acts described in the specification and equivalents thereof.
The following is a quotation of pre-AIA 35 U.S.C. 112, sixth paragraph:
An element in a claim for a combination may be expressed as a means or step for performing a specified function without the recital of structure, material, or acts in support thereof, and such claim shall be construed to cover the corresponding structure, material, or acts described in the specification and equivalents thereof.
The claims in this application are given their broadest reasonable interpretation using the plain meaning of the claim language in light of the specification as it would be understood by one of ordinary skill in the art. The broadest reasonable interpretation of a claim element (also commonly referred to as a claim limitation) is limited by the description in the specification when 35 U.S.C. 112(f) or pre-AIA 35 U.S.C. 112, sixth paragraph, is invoked.
As explained in MPEP § 2181, subsection I, claim limitations that meet the following three-prong test will be interpreted under 35 U.S.C. 112(f) or pre-AIA 35 U.S.C. 112, sixth paragraph:
(A) the claim limitation uses the term “means” or “step” or a term used as a substitute for “means” that is a generic placeholder (also called a nonce term or a non-structural term having no specific structural meaning) for performing the claimed function;
(B) the term “means” or “step” or the generic placeholder is modified by functional language, typically, but not always linked by the transition word “for” (e.g., “means for”) or another linking word or phrase, such as “configured to” or “so that”; and
(C) the term “means” or “step” or the generic placeholder is not modified by sufficient structure, material, or acts for performing the claimed function.
Use of the word “means” (or “step”) in a claim with functional language creates a rebuttable presumption that the claim limitation is to be treated in accordance with 35 U.S.C. 112(f) or pre-AIA 35 U.S.C. 112, sixth paragraph. The presumption that the claim limitation is interpreted under 35 U.S.C. 112(f) or pre-AIA 35 U.S.C. 112, sixth paragraph, is rebutted when the claim limitation recites sufficient structure, material, or acts to entirely perform the recited function.
Absence of the word “means” (or “step”) in a claim creates a rebuttable presumption that the claim limitation is not to be treated in accordance with 35 U.S.C. 112(f) or pre-AIA 35 U.S.C. 112, sixth paragraph. The presumption that the claim limitation is not interpreted under 35 U.S.C. 112(f) or pre-AIA 35 U.S.C. 112, sixth paragraph, is rebutted when the claim limitation recites function without reciting sufficient structure, material or acts to entirely perform the recited function.
Claim limitations in this application that use the word “means” (or “step”) are being interpreted under 35 U.S.C. 112(f) or pre-AIA 35 U.S.C. 112, sixth paragraph, except as otherwise indicated in an Office action. Conversely, claim limitations in this application that do not use the word “means” (or “step”) are not being interpreted under 35 U.S.C. 112(f) or pre-AIA 35 U.S.C. 112, sixth paragraph, except as otherwise indicated in an Office action.
This application includes one or more claim limitations that do not use the word “means,” but are nonetheless being interpreted under 35 U.S.C. 112(f) or pre-AIA 35 U.S.C. 112, sixth paragraph, because the claim limitation(s) uses a generic placeholder that is coupled with functional language without reciting sufficient structure to perform the recited function and the generic placeholder is not preceded by a structural modifier. Such claim limitation(s) is/are: “input terminal” in claim 20.
Because this/these claim limitation(s) is/are being interpreted under 35 U.S.C. 112(f) or pre-AIA 35 U.S.C. 112, sixth paragraph, it/they is/are being interpreted to cover the corresponding structure described in the specification as performing the claimed function, and equivalents thereof. Specifically, the “input terminal” is interpreted as a processor as per the specification [0239].
If applicant does not intend to have this/these limitation(s) interpreted under 35 U.S.C. 112(f) or pre-AIA 35 U.S.C. 112, sixth paragraph, applicant may: (1) amend the claim limitation(s) to avoid it/them being interpreted under 35 U.S.C. 112(f) or pre-AIA 35 U.S.C. 112, sixth paragraph (e.g., by reciting sufficient structure to perform the claimed function); or (2) present a sufficient showing that the claim limitation(s) recite(s) sufficient structure to perform the claimed function so as to avoid it/them being interpreted under 35 U.S.C. 112(f) or pre-AIA 35 U.S.C. 112, sixth paragraph.
Claim Rejections - 35 USC § 102
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 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.
(a)(2) the claimed invention was described in a patent issued under section 151, or in an application for patent published or deemed published under section 122(b), in which the patent or application, as the case may be, names another inventor and was effectively filed before the effective filing date of the claimed invention.
Claims 1, 17 and 20 are rejected under 35 U.S.C. 102(a)(1) as being anticipated by Zopf et al. (US Patent Pub. No. 20210350798 A1), hereinafter Zopf.
Regarding claims 1 and 20, Zopf teaches a method of speech recognition by a device and an apparatus (Zopf in [0022] teaches methods, and systems for detecting a wake word from uttered speech),
the method comprising:
[claim 20 only] an input terminal configured to receive an audio signal from one or more microphones; and a processing system configured to (Zopf in [0030] teaches using a microphone array and a device which processes audio, which are coupled together):
receiving a target phrase for recognition by a speech recognition model (Zopf in [0023] teaches using an audio processing device to receive sound waves from an audio pattern source, and in [0002] teaches using speech recognition systems for wake word detection);
analyzing a sequence of acoustic units representative of the target phrase when the target phrase is spoken to generate offline analysis data (Zopf in [0023] teaches the wake-words are trained offline, and in [0023] teaches using an audio processing device to receive sound waves from an audio pattern source, and in [0040] teaches using a phoneme-based training approach, wherein individual phonemes are trained offline to create a phoneme database);
constructing the speech recognition model based on the offline analysis data to decode speech signals of the target phrase according to the acoustic units (Zopf in [0022] teaches set of utterances may be analyzed by the first model to determine a second model, the second model comprising a training state sequence for each of the set of training utterances, and wherein each training state sequence indicates a likely state for each time interval of a corresponding training utterance, and in [0039] teaches in a standard training approach, dozens to hundreds of utterances of the wake word are used in offline training to produce the final word model);
and
processing speech based on the speech recognition model to detect a presence of the target phrase (Zopf in [0022] teaches methods, and systems for detecting a wake word from uttered speech using a first and second model).
Regarding claim 17, Zopf teaches the method of claim 1.
Zopf teaches the target phrase.
Zopf further teaches
wherein the target phrase comprises at least one of:
a wake-word spoken to address the device (Zopf in [0036] teaches using a buffer which may act as a channel to pass through the continuously sampled audio data comprising the wake phrase and the command or query);
a simple command spoken following the wake-word, wherein the simple command includes one or more words (Zopf in [0036] teaches using a buffer which may act as a channel to pass through the continuously sampled audio data comprising the wake phrase and the command or query);
a compound command spoken following the wake-word, wherein the compound command includes a common sub-command and a second sub-command unique to each compound command;
a number and an associated unit spoken following the wake-word;
or
a complex command spoken following the wake-word, wherein the complex command includes a combination of any one of the simple command, the compound command, and the number and the associated unit.
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.
Claim 2 is rejected under 35 U.S.C. 103 as being unpatentable over Zopf, in view of Fastow et al. (US Patent No. 9142209 B2), hereinafter Fastow.
Regarding claim 2, Zopf teaches the method of claim 1.
Zopf teaches the speech recognition model, the target phrase, and the acoustic units.
Zopf does not teach, however Fastow teaches
wherein processing speech based on the speech recognition model comprises:
determining from the [speech recognition model] a model likelihood score representing a likelihood of the presence of the [target phrase] based on an observed sequence of [acoustic units] decoded from the speech (Fastow in [col 1 ln 65 – col 2 ln 10] teaches using probability scores which are computed for each senone in the HMMs being considered, and a cumulative probability score is successively computed for each of the possible senone sequences as the Viterbi algorithm analyzes the sequential observation vectors).
Fastow is considered to be analogous to the claimed invention because it is in the same field of phoneme analysis. Therefore, it would have been obvious to someone of ordinary skill in the art before the effective filing date of the claimed invention to have modified Zopf further in view of Fastow to allow for using probability scores. Motivation to do so would allow for a plurality of data-pattern analysis units which allow for simultaneous processing of multiple data streams to allow real-time to near real-time analysis (Fastow [col 4 ln 50-61]).
Claim 3 is rejected under 35 U.S.C. 103 as being unpatentable over Zopf, in view of Fastow, in view of Bapat et al. (US Patent No. 9373321 B2), hereinafter Bapat.
Regarding claim 3, Zopf, as modified above, teaches the method of claim 2.
Zopf, as modified above, teaches the speech recognition model, the target phrase, the offline analysis data, and the sequence of acoustic units.
Zopf, as modified above, does not teach, however Bapat teaches
wherein processing speech based on the speech recognition model further comprises:
modifying the model likelihood score based on the observed [sequence of acoustic units] and the [offline analysis data] to determine the presence of the [target phrase] (Bapat in [col 7 ln 15-30] teaches using a confidence score calculation which refers to a confidence level in which a WUW speech recognizer (e.g., WUW processing engine 120 of FIG. 1) can accurately decode the one or more wake-up words, and the confidence score can be, for example, defined as a difference between log-likelihoods of the keyword and non-keyword models or a combination of statistics values of the likelihood of wake-up words in the keyword and non-keyword models).
Bapat is considered to be analogous to the claimed invention because it is in the same field of phoneme analysis. Therefore, it would have been obvious to someone of ordinary skill in the art before the effective filing date of the claimed invention to have modified Zopf, as modified above, further in view of Bapat to allow for using a confidence score which can be, for example, defined as a difference between log-likelihoods of the keyword and non-keyword models. Motivation to do so would allow for a television user (e.g., an end user) to modify or change the pre-set or defined wake-up words, while meeting a predetermined strength associated with the wake-up words, in an embodiment, the predetermined strength is based on the WUW speech recognizer achieving high accuracy—e.g., approximate to or at 100% accuracy—when decoding the modified or changed wake-up words (Bapat [col 3 ln 4-17]).
Claims 4 and 9 are rejected under 35 U.S.C. 103 as being unpatentable over Zopf, in view of Lopatka et al. (US Patent Pub. No. 20190042881 A1), hereinafter Lopatka.
Regarding claim 4, Zopf teaches the method of claim 1.
Zopf teaches the speech recognition model, and the sequence of acoustic units.
Zopf does not teach, however Lopatka teaches
wherein the [speech recognition model] comprises a sequence of decoding states (Lopatka in [0026, Fig. 6] teaches using a sequence decoder circuit),
wherein each decoding state of the sequence of decoding states models each [acoustic unit of the sequence of acoustic units] (Lopatka in [0026, Fig. 6] teaches using a sequence decoder circuit to predict acoustic events),
and
wherein analyzing the [sequence of acoustic units] comprises: determining an order of transitioning through the sequence of decoding states based on time (Lopatka in [0026, Fig. 6] teaches the acoustic event may be detected when the sequence decoder circuit has traversed states one through three in the correct order, with satisfactory classifier output scores, in some embodiments, an event detection may also require that the sequence decoder circuit remains in each of the three states for a selected time duration, and for example, the durations of each of the subparts of a given acoustic event may be known or anticipated to lie within a specific range of time duration values).
Lopatka is considered to be analogous to the claimed invention because it is in the same field of phoneme analysis. Therefore, it would have been obvious to someone of ordinary skill in the art before the effective filing date of the claimed invention to have modified Zopf further in view of Lopatka to allow for using a sequence decoder circuit. Motivation to do so would allow for improved acoustic event detection based on the modelling of event subparts with a finer level of granularity, which improves the discriminative power of the classifier, compared to existing systems that attempt to recognize the entirety of the event (Lopatka [0016]).
Regarding claim 9, Zopf teaches the method of claim 1.
Zopf teaches the speech recognition model, the offline analysis data, the sequence of decoding states, and the sequence of acoustic units.
Zopf further teaches
wherein the offline analysis data comprises:
one or more acoustically similar [acoustic units] to an [acoustic unit] modeled by a decoding state (Zopf in [0038] teaches wherein whole word models for each wake word are trained using any appropriate algorithm such as the maximum-likelihood forward-backward algorithm. Gaussian observation models are determined by using either diagonal or full covariance structure, with diagonal structure the predominant approach. The word models may comprise a “left-to-right” linear sequence of states with e.g., approximately three states per phoneme. The observation vectors O are obtained through a front end spectral analysis, with Mel-frequency cepstral coefficients (MFCCs) and their derivatives being the most common. In a standard training approach, dozens to hundreds of utterances of the wake word are used in offline training to produce the final word model as depicted in FIG. 3A. During decoding, a user may wish to calculate the probability of the observation sequence, O=O.sub.1O.sub.2 . . . O.sub.T given the model), i.e. P(O|λ). Any appropriate algorithm may be used to perform this calculation, such as the Viterbi algorithm),
wherein the acoustically similar [acoustic units] are associated with probability estimates of a presence of the acoustically similar [acoustic units] when the decoding state identifies the [acoustic unit] modeled by the decoding state as a most likely [acoustic unit] (Zopf in [0038] teaches wherein whole word models for each wake word are trained using any appropriate algorithm such as the maximum-likelihood forward-backward algorithm. Gaussian observation models are determined by using either diagonal or full covariance structure, with diagonal structure the predominant approach. The word models may comprise a “left-to-right” linear sequence of states with e.g., approximately three states per phoneme. The observation vectors O are obtained through a front end spectral analysis, with Mel-frequency cepstral coefficients (MFCCs) and their derivatives being the most common. In a standard training approach, dozens to hundreds of utterances of the wake word are used in offline training to produce the final word model as depicted in FIG. 3A. During decoding, a user may wish to calculate the probability of the observation sequence, O=O.sub.1O.sub.2 . . . O.sub.T given the model), i.e. P(O|λ). Any appropriate algorithm may be used to perform this calculation, such as the Viterbi algorithm).
Zopf does not teach, however Lopatka teaches
wherein the speech recognition model comprises a sequence of decoding states (Lopatka in [0026, Fig. 6] teaches using a sequence decoder circuit),
wherein each decoding state of the sequence of decoding states models each [acoustic unit] of the [sequence of acoustic units] (Lopatka in [0026, Fig. 6] teaches using a sequence decoder circuit to predict acoustic events).
Lopatka is considered to be analogous to the claimed invention because it is in the same field of phoneme analysis. Therefore, it would have been obvious to someone of ordinary skill in the art before the effective filing date of the claimed invention to have modified Zopf further in view of Lopatka to allow for using a sequence decoder circuit. Motivation to do so would allow for improved acoustic event detection based on the modelling of event subparts with a finer level of granularity, which improves the discriminative power of the classifier, compared to existing systems that attempt to recognize the entirety of the event (Lopatka [0016]).
Claims 5, 7, 18 and 19 are rejected under 35 U.S.C. 103 as being unpatentable over Zopf, in view of Nagano et al. (US Patent Pub. No. 20120197644 A1), hereinafter Nagano.
Regarding claim 5, Zopf teaches the method of claim 1.
Zopf teaches the target phrase, and the offline analysis data.
Zopf further teaches
wherein the [target phrase] comprises a plurality of words (Zopf in [0038] teaches determine whether the contiguous audio data output by the combiner includes a wake word or wake phrase).
Zopf does not teach, however Nagano teaches
wherein the [offline analysis data] comprises an expected length in time of the [target phrase] (Nagano in [0040] teaches performing sound analysis on a speech section separated by pauses within a speech spectrum included in speech data having a particular time length to derive such features as temporal length of a word or phrase, fundamental frequency, magnitude, and cepstrum)
and
an expected length in time of each of the plurality of words when the [target phrase] is spoken (Nagano in [0040] teaches performing sound analysis on a speech section separated by pauses within a speech spectrum included in speech data having a particular time length to derive such features as temporal length of a word or phrase, fundamental frequency, magnitude, and cepstrum).
Nagano is considered to be analogous to the claimed invention because it is in the same field of speech analysis. Therefore, it would have been obvious to someone of ordinary skill in the art before the effective filing date of the claimed invention to have modified Zopf further in view of Nagano to allow for performing sound analysis on a speech. Motivation to do so would allow for extraction of characteristic words in accordance with the speaker, such as an employee, allowing efficient extraction of key phrases reflecting a subtle change in mental attitude that cannot be identified from text alone, including a result of speech recognition (Nagano [0067]).
Regarding claim 7, Zopf teaches the method of claim 1.
Zopf teaches the offline analysis data, and the sequence of acoustic units.
Zopf does not teach, however Nagano teaches
wherein the [offline analysis data] comprises an expected length in time of each of the [acoustic units] in the [sequence of acoustic units] (Nagano in [0040] teaches performing sound analysis on a speech section separated by pauses within a speech spectrum included in speech data having a particular time length to derive such features as temporal length of a word or phrase, fundamental frequency, magnitude, and cepstrum).
Nagano is considered to be analogous to the claimed invention because it is in the same field of speech analysis. Therefore, it would have been obvious to someone of ordinary skill in the art before the effective filing date of the claimed invention to have modified Zopf further in view of Nagano to allow for performing sound analysis on a speech. Motivation to do so would allow for extraction of characteristic words in accordance with the speaker, such as an employee, allowing efficient extraction of key phrases reflecting a subtle change in mental attitude that cannot be identified from text alone, including a result of speech recognition (Nagano [0067]).
Regarding claim 18, Zopf teaches the method of claim 17.
Zopf, as modified above, teaches the speech recognition model, the offline analysis data, the sequence decoding model, the sequence of decoding states, and the sequence of acoustic units.
Zopf further teaches
wherein constructing the [speech recognition model] based on the [offline analysis data] comprises:
constructing a [sequence decoding model] based on a [sequence of acoustic units] of the wake-word followed by a [sequence of acoustic units] of a command (Zopf in [0036] teaches using a buffer which may act as a channel to pass through the continuously sampled audio data comprising the wake phrase and the command or query),
wherein the [sequence decoding model] models
a first [sequence of states] corresponding to the [sequence of acoustic units] of the wake-word (Zopf in [0036] teaches using a buffer which may act as a channel to pass through the continuously sampled audio data comprising the wake phrase and the command or query)
and
a second [sequence of states] corresponding to the [sequence of acoustic units] of the command (Zopf in [0036] teaches using a buffer which may act as a channel to pass through the continuously sampled audio data comprising the wake phrase and the command or query),
and
wherein a state of the first [sequence of states] corresponding to a last [acoustic unit] of the [wake-word] also models [a gap between the wake-word and the command] (Zopf in [0036] teaches using a buffer which may act as a channel to pass through the continuously sampled audio data comprising the wake phrase and the command or query).
Zopf, as modified above, does not teach, however Nagano teaches
a gap between the [wake-word and the command] (Nagano in [0046] teaches identifying a particular word or phrase by detecting the presence of pauses, or silent sections, that are present before and after the word or phrase in order to characterize a conversation, and extracts words for use in emotion analysis, and a pause as called herein can be defined as a section in which silence is recorded for a certain length on both sides of a speech spectrum, as shown by a rectangular area 400 in the speech data).
Nagano is considered to be analogous to the claimed invention because it is in the same field of speech analysis. Therefore, it would have been obvious to someone of ordinary skill in the art before the effective filing date of the claimed invention to have modified Zopf further in view of Nagano to allow for performing gap analysis on a speech. Motivation to do so would allow for extraction of characteristic words in accordance with the speaker, such as an employee, allowing efficient extraction of key phrases reflecting a subtle change in mental attitude that cannot be identified from text alone, including a result of speech recognition (Nagano [0067]).
Regarding claim 19, Zopf teaches the method of claim 17.
Zopf, as modified above, teaches the speech recognition model, the offline analysis data, the sequence decoding model, the sequence of decoding states, the target phrase and the sequence of acoustic units.
Zopf further teaches
wherein constructing the [speech recognition model] based on the [offline analysis data] comprises:
constructing a [sequence decoding model] based on a concatenation of [sequences of acoustic units] of a plurality of words of a command (Zopf in [0042] teaches using a speaker-independent concatenated-phoneme based model, and in [0036] teaches using a buffer which may act as a channel to pass through the continuously sampled audio data comprising the wake phrase and the command or query),
wherein the [sequence decoding model] includes
a first [sequence of states] modeling a [sequence of acoustic units] of a first word (Zopf in [0036] teaches using a buffer which may act as a channel to pass through the continuously sampled audio data comprising the wake phrase and the command or query),
[an silence state modeling a gap] between the first word and a second word of the command (Zopf in [0036] teaches using a buffer which may act as a channel to pass through the continuously sampled audio data comprising the wake phrase and the command or query),
and
a second [sequence of states] modeling a [sequence of acoustic units] of the second word (Zopf in [0036] teaches using a buffer which may act as a channel to pass through the continuously sampled audio data comprising the wake phrase and the command or query),
and
wherein [the silence state also models] a last [acoustic unit] of the first word and a first [acoustic unit] of the second word (Zopf in [0036] teaches using a buffer which may act as a channel to pass through the continuously sampled audio data comprising the wake phrase and the command or query).
Zopf, as modified above, does not teach, however Nagano teaches
a gap between the [wake-word and the command] (Nagano in [0046] teaches identifying a particular word or phrase by detecting the presence of pauses, or silent sections, that are present before and after the word or phrase in order to characterize a conversation, and extracts words for use in emotion analysis, and a pause as called herein can be defined as a section in which silence is recorded for a certain length on both sides of a speech spectrum, as shown by a rectangular area 400 in the speech data);
an silence state modeling a gap between [the first word and a second word of the command] (Nagano in [0046] teaches identifying a particular word or phrase by detecting the presence of pauses, or silent sections, that are present before and after the word or phrase in order to characterize a conversation, and extracts words for use in emotion analysis, and a pause as called herein can be defined as a section in which silence is recorded for a certain length on both sides of a speech spectrum, as shown by a rectangular area 400 in the speech data).
Nagano is considered to be analogous to the claimed invention because it is in the same field of speech analysis. Therefore, it would have been obvious to someone of ordinary skill in the art before the effective filing date of the claimed invention to have modified Zopf further in view of Nagano to allow for performing gap analysis on a speech. Motivation to do so would allow for extraction of characteristic words in accordance with the speaker, such as an employee, allowing efficient extraction of key phrases reflecting a subtle change in mental attitude that cannot be identified from text alone, including a result of speech recognition (Nagano [0067]).
Claim 6 is rejected under 35 U.S.C. 103 as being unpatentable over Zopf, in view of Fastow, in view of Lopatka.
Regarding claim 6, Zopf teaches the method of claim 1.
Zopf teaches the target phrase, the sequence of acoustic units, and the speech recognition model.
Zopf does not teach, however Fastow teaches
wherein processing speech based on the [speech recognition model] comprises determining a likelihood of the presence of the [target phrase] (Fastow in [col 1 ln 65 – col 2 ln 10] teaches using probability scores which are computed for each senone in the HMMs being considered, and a cumulative probability score is successively computed for each of the possible senone sequences as the Viterbi algorithm analyzes the sequential observation vectors).
Fastow is considered to be analogous to the claimed invention because it is in the same field of phoneme analysis. Therefore, it would have been obvious to someone of ordinary skill in the art before the effective filing date of the claimed invention to have modified Zopf further in view of Fastow to allow for using probability scores. Motivation to do so would allow for a plurality of data-pattern analysis units which allow for simultaneous processing of multiple data streams to allow real-time to near real-time analysis (Fastow [col 4 ln 50-61]).
Zopf, as modified above, does not teach, however Lopatka teaches
wherein the speech recognition model comprises a sequence of states (Lopatka in [0026, Fig. 6] teaches using a sequence decoder circuit),
wherein each state of the [sequence of states] models each [acoustic unit] of the [sequence of acoustic units] (Lopatka in [0026, Fig. 6] teaches using a sequence decoder circuit to predict acoustic events),
and
based on an order of transitions between states of the [sequence of states] when the [target phrase] is spoken (Lopatka in [0026, Fig. 6] teaches using a sequence decoder circuit, with acoustic event may be detected when the sequence decoder circuit has traversed states one through three in the correct order).
Lopatka is considered to be analogous to the claimed invention because it is in the same field of phoneme analysis. Therefore, it would have been obvious to someone of ordinary skill in the art before the effective filing date of the claimed invention to have modified Zopf, as modified above, further in view of Lopatka to allow for using a sequence decoder circuit. Motivation to do so would allow for improved acoustic event detection based on the modelling of event subparts with a finer level of granularity, which improves the discriminative power of the classifier, compared to existing systems that attempt to recognize the entirety of the event (Lopatka [0016]).
Claim 8 is rejected under 35 U.S.C. 103 as being unpatentable over Zopf, in view of Nagano, in view of Lopatka.
Regarding claim 8, Zopf, as modified above, teaches the method of claim 7.
Zopf, as modified above, teaches the speech recognition model, the sequence of decoding states, and the sequence of acoustic units.
Zopf, as modified above, does not teach, however Lopatka teaches
wherein the [speech recognition model] comprises a sequence of decoding states (Lopatka in [0026, Fig. 6] teaches using a sequence decoder circuit),
wherein each decoding state of the sequence of decoding states models each [acoustic unit] of the [sequence of acoustic units] (Lopatka in [0026, Fig. 6] teaches using a sequence decoder circuit to predict acoustic events),
and
wherein the expected length in time of each of the [acoustic units] comprises an expected length in time the [speech recognition model] stays in [each of the decoding states] when decoding speech signals of the [target phrase] (Lopatka in [0026, Fig. 6] teaches the acoustic event may be detected when the sequence decoder circuit has traversed states one through three in the correct order, with satisfactory classifier output scores, in some embodiments, an event detection may also require that the sequence decoder circuit remains in each of the three states for a selected time duration, and for example, the durations of each of the subparts of a given acoustic event may be known or anticipated to lie within a specific range of time duration values).
Lopatka is considered to be analogous to the claimed invention because it is in the same field of phoneme analysis. Therefore, it would have been obvious to someone of ordinary skill in the art before the effective filing date of the claimed invention to have modified Zopf further in view of Lopatka to allow for using a sequence decoder circuit. Motivation to do so would allow for improved acoustic event detection based on the modelling of event subparts with a finer level of granularity, which improves the discriminative power of the classifier, compared to existing systems that attempt to recognize the entirety of the event (Lopatka [0016]).
Claim 10 is rejected under 35 U.S.C. 103 as being unpatentable over Zopf, in view of Nagano, in view of Lopatka, in view of Amadori et al. (US Patent Pub. No. 20250173911 A1), hereinafter Amadori.
Regarding claim 10, Zopf teaches the method of claim 1.
Zopf teaches the speech recognition model, the offline analysis data, the sequence of decoding states, and the sequence of acoustic units.
Zopf does not teach, however Nagano teaches
wherein constructing the [speech recognition model] based on the [offline analysis data] comprises:
constructing a [sequence decoding model] based on the [sequence of acoustic units] (Zopf in [0022] teaches using a processing device which may determine a first model for wake word recognition based on a likelihood ratio using a set of training utterances. The set of utterances may be analyzed by the first model to determine a second model, the second model comprising a training state sequence for each of the set of training utterances, and wherein each training state sequence indicates a likely state for each time interval of a corresponding training utterance. A determination of whether a detected utterance corresponds to the wake word may be based on a concatenation of the first model and the second model. More specifically, the processing device may measure a distance between each training state sequence and a state sequence of the detected utterance to generate a set of distances and a minimum distance among the set of distances),
Zopf does not teach, however Lopatka teaches
wherein the sequence decoding model includes a sequence of states (Lopatka in [0026, Fig. 6] teaches using a sequence decoder circuit),
and
wherein each state of the sequence of states models each [acoustic unit] of the [sequence of acoustic units] (Lopatka in [0026, Fig. 6] teaches using a sequence decoder circuit to predict acoustic events).
Lopatka is considered to be analogous to the claimed invention because it is in the same field of phoneme analysis. Therefore, it would have been obvious to someone of ordinary skill in the art before the effective filing date of the claimed invention to have modified Zopf further in view of Lopatka to allow for using a sequence decoder circuit. Motivation to do so would allow for improved acoustic event detection based on the modelling of event subparts with a finer level of granularity, which improves the discriminative power of the classifier, compared to existing systems that attempt to recognize the entirety of the event (Lopatka [0016]).
Zopf, as modified above, does not teach, however Amadori teaches
constructing a decoding compensation model based on the [offline analysis data] to modify a decoding output of the [sequence decoding model] (Amadori in [0068]teaches the decoder model is operable to update one or more of the parameters of the decoder model to update the decoder model to compensate for differences between the predicted caption data for the reconstructed image and the received caption data).
Amadori is considered to be analogous to the claimed invention because it is in the same field of updating a model to compensate for differences. Therefore, it would have been obvious to someone of ordinary skill in the art before the effective filing date of the claimed invention to have modified Zopf, as modified above, further in view of Amadori to allow for using a decoder model to update the decoder model parameters to compensate for differences. Motivation to do so would allow for encoded data having an associated image quality and associated caption data can be provided to a decoder, and the caption data indicative of the language-based description can assist in allowing an improvement of the quality of the generated reconstructed image relative to a comparative case in which the encoded data is provided on its own (Amadori [0038]).
Claims 11-13 are rejected under 35 U.S.C. 103 as being unpatentable over Zopf, in view of Nagano, in view of Lopatka, in view of Amadori, in view of Enzinger et al. (US Patent Pub. No. 20210193174 A1), hereinafter Enzinger.
Regarding claim 11, Zopf, as modified above, teaches the method of claim 10.
Zopf, as modified above, teaches the speech recognition model, the sequence decoding model, the sequence of decoding states, the decoding compensation model, the target phrase and the acoustic units.
Zopf, as modified above, does not teach, however Enzinger teaches
wherein the [sequence decoding model] decodes a most likely path through the [sequence of states] when processing speech (Enzinger in [0009] teaches the speech decoder can incorporate a weighted finite state transducer (WFST) and can generate a lattice of alternative word sequences including a path that is the most likely sequence of words),
and
wherein the [decoding compensation model] compares transitions through the [sequence of states] of the most likely path with expected transitions through the [sequence of states] when the [speech recognition model] processes [acoustic units] of the [target phrase] (Enzinger in [0009] teaches the real-time notification can be issued responsive to a determination that the measure of likelihood exceeds a threshold).
Enzinger is considered to be analogous to the claimed invention because it is in the same field of speech processing. Therefore, it would have been obvious to someone of ordinary skill in the art before the effective filing date of the claimed invention to have modified Zopf, as modified above, further in view of Enzinger to allow for using a speech decoder which can incorporate a weighted finite state transducer (WFST) and can generate a lattice of alternative word sequences including a path that is the most likely sequence of words. Motivation to do so would allow for detection of tampering, suspicious content, or spoofing in an incoming voice signal, in varying combinations, with alerts (Enzinger [0004]).
Regarding claim 12, Zopf, as modified above, teaches the method of claim 11.
Zopf, as modified above, teaches the sequence of states, the target phrase, and the acoustic units.
Zopf further teaches
wherein the expected transitions through the [sequence of states] comprises at least one of:
a ratio between an expected length in time of a word in the [target phrase] and an expected total length in time of the [target phrase] when the [target phrase] is spoken;
an expected transition of 1 state through the [sequence of states] when the [target phrase] is spoken (Zopf in [0042] teaches using state combining which involves the merging of two or more similar or identical states with a very high likelihood of transition between them);
an expected length in time in each state of the [sequence of states] when the [target phrase] is spoken;
or
acoustically similar [acoustic units] to an [acoustic unit] that is modeled by each state of the [sequence of states],
wherein each of the acoustically similar [acoustic units] is associated with a probability estimate of a detection when the [acoustic unit] is modeled by a [corresponding state].
Regarding claim 13, Zopf, as modified above, teaches the method of claim 12.
Zopf, as modified above, teaches the speech recognition model, the sequence decoding model, the sequence of decoding states, the decoding compensation model, the target phrase and the acoustic units.
Zopf, as modified above, does not teach, however Enzinger teaches
wherein processing speech based on the speech recognition model comprises at least one of:
comparing a ratio of an observed length in time of a word in the speech and an observed total length in time of the speech when transitioning through the [sequence of states] of the most likely path with the ratio between an expected length in time of a word in the [target phrase] and an expected total length in time of the [target phrase] to generate a word-ratio penalty;
comparing observed state jumps when transitioning through the [sequence of states] of the most likely path with the expected transition of 1 state for the [target phrase] to generate a state jump penalty (Enzinger in [0079] teaches using a “weighted finite-state transducer” (WFST) is a finite automaton having state transitions from input to output states sometimes denoted by respective symbols, and additionally or alternatively to having weights on the input and output states, weights can be assigned to the transitions, the WFST can be applied for speech decoding, and in [0009] teaches the speech decoder can incorporate a weighted finite state transducer (WFST) and can generate a lattice of alternative word sequences including a path that is the most likely sequence of words);
comparing an observed length in time in each state when transitioning through the sequence of states of the most likely path with the expected length in time in each state of the target phrase to generate a state walk penalty;
or
comparing a probability estimate of a most likely acoustic unit modeled by each state when transitioning through the sequence of states of the most likely path with probability estimates for the acoustically similar acoustic units and the acoustic unit modeled by each state for the expected transitions of the target phrase to generate a top-1 penalty.
Enzinger is considered to be analogous to the claimed invention because it is in the same field of speech processing. Therefore, it would have been obvious to someone of ordinary skill in the art before the effective filing date of the claimed invention to have modified Zopf, as modified above, further in view of Enzinger to allow for using a speech decoder which can incorporate a weighted finite state transducer (WFST). Motivation to do so would allow for detection of tampering, suspicious content, or spoofing in an incoming voice signal, in varying combinations, with alerts (Enzinger [0004]).
Allowable Subject Matter
Claims 14-16 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.
Conclusion
Any inquiry concerning this communication or earlier communications from the examiner should be directed to PAUL J. MUELLER whose telephone number is (571)272-1875. The examiner can normally be reached M-F 9:00am-5:00pm (Eastern).
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, Daniel C. Washburn can be reached at 571-272-5551. 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.
PAUL MUELLER
Examiner
Art Unit 2657
/PAUL J. MUELLER/Examiner, Art Unit 2657
Prosecution Insights