Notice of Pre-AIA Status
The present application is being examined under the pre-AIA first to invent provisions.
DETAILED ACTION
This office action is in response to application 19014095, which was filed 01/08/25 and is a continuation of Application 16900857, now U.S. Patent 12223963, which is a continuation of Application 15603257, now abandoned, which is a continuation of Application 15085944, now U.S. Patent 9691390, which is a continuation of Application 14621024, now U.S. Patent 9330669, which is a continuation of Application 13530101, now U.S. Patent 8972263. Claims 1-20 are pending in the application and have been considered.
Claim Objections
In order to expedite prosecution, the examiner will assume, for examination purposes, that Applicant intended dependent claim 20 to depend on independent claim 19 instead of claim 18. Appropriate correction is required.
Claim Rejections - 35 USC § 103
The following is a quotation of pre-AIA 35 U.S.C. 103(a) which forms the basis for all obviousness rejections set forth in this Office action:
(a) A patent may not be obtained though the invention is not identically disclosed or described as set forth in section 102 of this title, if the differences between the subject matter sought to be patented and the prior art are such that the subject matter as a whole would have been obvious at the time the invention was made to a person having ordinary skill in the art to which said subject matter pertains. Patentability shall not be negatived by the manner in which the invention was made.
Claims 1, 11, 18, and 19 are rejected under pre-AIA 35 U.S.C. 103(a) as being unpatentable over Burns et al. (US 20070156412) in view of Thelen et al. (US 6487534).
Consider claim 1, Burns discloses a method comprising:
transmitting, via a communication link, a spoken query from a host device to a remote server having a speech recognition system (user speaks a command such as a query to “find movie” to device 10, which wireless sends the recorded voice command file to server computer 20, [0012], [0021], Fig. 1);
performing speech recognition on the spoken query (server computer 20 performs speech recognition 40 on the recorded voice command file 30, [0013], Fig. 1);
obtaining a speech recognition transcription and a corresponding score from the remote server (server 200 performs speech recognition to result in creation of an interpreted voice command file with confidence level for the transcribed voice command, [0025]); and,
in dependence on the speech recognition transcription and the corresponding score, issuing, by a control module on the host device, a command to control the host device to perform one or more operations (if the confidence level resulting from the speech recognition is at or above the critical confidence level, the application is automatically created using the interpreted voice command, resulting in creation of machine readable command file, which is transmitted back to communication device for execution of the command, [0026]-[0028]).
Burns does not specifically mention accessing, by a speech recognition module, a language context comprising: (i) a first module comprising a vocabulary that includes a set of words, and (ii) a second module comprising a language model.
Thelen discloses accessing, by a speech recognition module, a language context comprising: (i) a first module comprising a vocabulary that includes a set of words, and (ii) a second module comprising a language model (speech recognition system 100, i.e. “speech recognition module”, includes unit matching subsystem 120, i.e. “language context”, which is made of word matching system 130 utilizing lexicon 134, i.e. “a first module”, and sentence level matching system based on language model, i.e. “a second module”, Col 4 lines 10-15, Col 5 lines 37-49, Fig 1).
It would have been obvious to one of ordinary skill in the art at the time of the invention to modify Burns by accessing, by a speech recognition module, a language context comprising: (i) a first module comprising a vocabulary that includes a set of words, and (ii) a second module comprising a language model as in Thelen in order place constraints on the speech recognition matching processes (Thelen, Col 5 lines 35-58), predictably improving accuracy and reducing system load, as suggested by Thelen (Col 1-2 lines 67-6). The cited references are analogous art in the field of speech recognition.
Consider claim 11, Burns discloses a non-transitory computer readable medium comprising code that, when executed by one or more processors (memory and processors executing code to operate communication device 10 and server computers 20, [0011], Fig 1), causes the processor to perform a method comprising:
transmitting, via a communication link, a spoken query from a host device to a remote server having a speech recognition system (user speaks a command such as a query to “find movie” to device 10, which wireless sends the recorded voice command file to server computer 20, [0012], [0021], Fig. 1);
performing speech recognition on the spoken query (server computer 20 performs speech recognition 40 on the recorded voice command file 30, [0013], Fig. 1);
obtaining a speech recognition transcription and a corresponding score from the remote server (server 200 performs speech recognition to result in creation of an interpreted voice command file with confidence level for the transcribed voice command, [0025]); and,
in dependence on the speech recognition transcription and the corresponding score, issuing, by a control module on the host device, a command to control the host device to perform one or more operations (if the confidence level resulting from the speech recognition is at or above the critical confidence level, the application is automatically created using the interpreted voice command, resulting in creation of machine readable command file, which is transmitted back to communication device for execution of the command, [0026]-[0028]).
Burns does not specifically mention accessing, by a speech recognition module, a language context comprising: (i) a first module comprising a vocabulary that includes a set of words, and (ii) a second module comprising a language model.
Thelen discloses accessing, by a speech recognition module, a language context comprising: (i) a first module comprising a vocabulary that includes a set of words, and (ii) a second module comprising a language model (speech recognition system 100, i.e. “speech recognition module”, includes unit matching subsystem 120, i.e. “language context”, which is made of word matching system 130 utilizing lexicon 134, i.e. “a first module”, and sentence level matching system based on language model, i.e. “a second module”, Col 4 lines 10-15, Col 5 lines 37-49, Fig 1).
It would have been obvious to one of ordinary skill in the art at the time of the invention to modify Burns by accessing, by a speech recognition module, a language context comprising: (i) a first module comprising a vocabulary that includes a set of words, and (ii) a second module comprising a language model as in Thelen for reasons similar to those for claim 1.
Consider claim 19, Burns discloses a speech recognition system including one or more processors coupled to memory, the memory being loaded with computer instructions to control a host device to perform one or more operations, the computer instructions, when executed on the one or more processors, causing the one or more processors to implement operations (memory and processors executing code to operate communication device 10 and server computers 20, [0011], Fig 1) comprising:
transmitting, via a communication link, a spoken query from a host device to a remote server having a speech recognition system (user speaks a command such as a query to “find movie” to device 10, which wireless sends the recorded voice command file to server computer 20, [0012], [0021], Fig. 1);
performing speech recognition on the spoken query (server computer 20 performs speech recognition 40 on the recorded voice command file 30, [0013], Fig. 1);
obtaining a speech recognition transcription and a corresponding score from the remote server (server 200 performs speech recognition to result in creation of an interpreted voice command file with confidence level for the transcribed voice command, [0025]); and,
in dependence on the speech recognition transcription and the corresponding score, issuing, by a control module on the host device, a command to control the host device to perform one or more operations (if the confidence level resulting from the speech recognition is at or above the critical confidence level, the application is automatically created using the interpreted voice command, resulting in creation of machine readable command file, which is transmitted back to communication device for execution of the command, [0026]-[0028]).
Burns does not specifically mention accessing, by a speech recognition module, a language context comprising: (i) a first module comprising a vocabulary that includes a set of words, and (ii) a second module comprising a language model.
Thelen discloses accessing, by a speech recognition module, a language context comprising: (i) a first module comprising a vocabulary that includes a set of words, and (ii) a second module comprising a language model (speech recognition system 100, i.e. “speech recognition module”, includes unit matching subsystem 120, i.e. “language context”, which is made of word matching system 130 utilizing lexicon 134, i.e. “a first module”, and sentence level matching system based on language model, i.e. “a second module”, Col 4 lines 10-15, Col 5 lines 37-49, Fig 1).
It would have been obvious to one of ordinary skill in the art at the time of the invention to modify Burns by accessing, by a speech recognition module, a language context comprising: (i) a first module comprising a vocabulary that includes a set of words, and (ii) a second module comprising a language model as in Thelen for reasons similar to those for claim 1.
Consider claim 18, Burns does not, but Thelen discloses the vocabulary comprises words and phrases, and the language model comprises a set of constraints on word sequences expressed as N-grams and grammars (lexicon includes words and phrases, and language model includes n-grams and grammar constraints, Col 4 lines 10-15, Col 5 lines 37-49, Fig 1).
It would have been obvious to one of ordinary skill in the art at the time of the invention to modify Burns such that the vocabulary comprises words and phrases, and the language model comprises a set of constraints on word sequences expressed as N-grams and grammars for reasons similar to those for claim 1.
Claim 2 is rejected under pre-AIA 35 U.S.C. 103(a) as being unpatentable over Burns et al. (US 20070156412) in view of Thelen et al. (US 6487534), in further view of Carraux et al. (US 20110288857).
Consider claim 2, Burns discloses the speech recognition transcription is obtained from the speech recognition system of the remote server (server 200 performs speech recognition to result in creation of an interpreted voice command file, [0025]).
Burns and Thelen do not specifically mention prior to an expiration of a timeout, with duration of the timeout being determined according to a latency of the communication link used to transmit the spoken query.
Carraux discloses prior to an expiration of a timeout, with duration of the timeout being determined according to a latency of the communication link used to transmit the spoken query (recognition result appears in output queue before timeout value 404c is reached, [0038], timeout value 404c chosen to be shorter than the HTTP timeout value, e.g. 100-500 ms based on HTTP call network latency, [0040], [0008]).
It would have been obvious to one of ordinary skill in the art at the time of the invention to modify Burns and Thelen such that the speech recognition transcription is obtained from the speech recognition system of the remote server prior to an expiration of a timeout, with duration of the timeout being determined according to a latency of the communication link used to transmit the spoken query in order to avoid results that appear sluggish, as suggested by Carraux ([0008]), particularly important in control and command applications, as noted by Carraux ([0008]). The cited references are analogous art in the field of speech recognition.
Claims 8 is rejected under pre-AIA 35 U.S.C. 103(a) as being unpatentable over Burns et al. (US 20070156412) in view of Thelen et al. (US 6487534), in further view of Kao et al. (US 6374222).
Consider claim 8, Burns and Thelen do not, but Kao discloses responsive to available memory resources for client vocabulary data being about to run out, performing a garbage collection operation (slots are freed for every 20ms input acoustic vector, i.e. updated, via garbage collection, Col 7 lines 23-26).
It would have been obvious to one of ordinary skill in the art at the time of the invention to modify the invention of Burns and Thelen such that responsive to available memory resources for client vocabulary data being about to run out, performing a garbage collection operation in order to reduce the search space that does not introduce error, as suggested by Kao (Col 2 lines 32-35). Doing so would have predictably resulted in increased speed for a recognition search that can be used independently of multi-pass search or lexical tree, as identified by Kao (Col 2 lines 32-35). The cited references are analogous art in the field of speech recognition.
Claims 9 is rejected under pre-AIA 35 U.S.C. 103(a) as being unpatentable over Burns et al. (US 20070156412) in view of Thelen et al. (US 6487534), in further view of Lloyd et al. (US 20120022874).
Consider claim 9, Burns and Thelen do not, but Lloyd discloses combining frequency and recency of use to determine a word frequency amortized over time; and choosing a priority of a non-permanent word using the word frequency amortized over time (affinity score based on frequency and recency, [0011]).
It would have been obvious to one of ordinary skill in the art at the time of the invention to modify the invention of Burns and Thelen by combining frequency and recency of use to determine a word frequency amortized over time; and choosing a priority of a non-permanent word using the word frequency amortized over time in order to avoid misrecognition, as suggested by Llloyd ([0003]). The cited references are analogous art in the field of speech recognition.
Claims 14-17 are rejected under pre-AIA 35 U.S.C. 103(a) as being unpatentable over Burns et al. (US 20070156412) in view of Thelen et al. (US 6487534), in further view of Josifovski (US 20070038450).
Consider claim 14, Burns and Thelen do not, but Josifovski discloses a language context includes phonetic strings (phoneme strings, [0116]).
It would have been obvious to one of ordinary skill in the art at the time of the invention to modify the invention of Burns and Thelen such that the language context includes phonetic strings in order to assist the user in retrieving the desired information from a database quickly and efficiently, as suggested by Josifovski ([0002]). Doing so would have predictably allowed searching of non-text databases such as video or audio databases, as identified by Josifovski ([0003]). The cited references are analogous art in the field of speech recognition.
Consider claim 15, Burns and Thelen do not, but Josifovski discloses the language context includes one or more phonetic strings per pronunciation of a word (phonetic transcription unit identifies a number of possible phoneme strings, [0030-0031]).
It would have been obvious to one of ordinary skill in the art at the time of the invention to modify the invention of Burns and Thelen such that the language context includes one or more phonetic strings per pronunciation of a word for reasons similar to those for claim 14.
Consider claim 16, Burns and Thelen do not, but Josifovski discloses the language context includes phonetic lattices (phoneme lattices, [0031]).
It would have been obvious to one of ordinary skill in the art at the time of the invention to modify the invention of Burns and Thelen such that the language context includes phonetic lattices for reasons similar to those for claim 14.
Consider claim 17, Burns and Thelen do not, but Josifovski discloses the language context includes one phonetic lattice per word of the set of words (generating a phoneme lattice for each word in the utterance, [0031]).
It would have been obvious to one of ordinary skill in the art at the time of the invention to modify the invention of Burns and Thelen such that the language context includes one phonetic lattice per word of the set of words for reasons similar to those for claim 14.
Allowable Subject Matter
Claims 3-7, 10, 12-13, and 20 are objected to as being dependent on a rejected base claim, but would be allowable if rewritten in independent form including all limitations of the base and any intervening claims.
Claim 3 recites: “….determining that the speech recognition transcription contains a word that is missing from a local vocabulary; and, responsive to the determination, issuing a request to the server to send a description of the missing word; receiving the description of the missing word; and updating the local vocabulary with the description of the missing word.”
Claim 7 recites: “…responsive to determining that the speech recognition transcription contains a word that is missing from a local vocabulary, issuing a request to the server to send a description of the missing word; performing local speech recognition on the spoken query, using a local vocabulary; and comparing the corresponding score from the remote server to a local speech recognition score, wherein the issuing of the request to the server to send the description of the missing word depends on the outcome of the comparison.”
Claim 10 recites: “…responsive to determining that the speech recognition transcription contains a word that is missing from a local vocabulary, issuing a request to the server to send a description of the missing word, wherein the description includes a phonetic lattice.”
Chikuri (US 20100076751) discloses a distributed speech recognition system in which the user utters a word such as “florist”. The system transmits local dictionary 2b and the audio to server, which recognizes “Florist” determines “Florist” is not included in second voice recognition dictionary 2b of local device, and transmits it to local device for updating vocabulary 2b via communication unit [0057-0061], Fig. 9. However, Chikuri does not disclose “… responsive to the determination, issuing a request to the server to send a description of the missing word; receiving the description of the missing word;”.
Loring et al. (US 6195641) describes a distributed recognition system in which when a user at a client adds a new word to the system, a universal vocabulary list is updated at the server and the server sends the new vocabulary along with pronunciation information to other clients in the work group, Col 5-6 lines 57-7, Fig. 1.
Cohen et al. (US 20070233487) describes a distributed speech recognition system in which a pronunciation and phoneme dictionary updates 423 and 434 update the vocabulary, and are received from external search server 201, Fig. 4, [0062], by adding new entries such as “Johnny Appleseed” to the dictionary, [0064].
However, Loring, Cohen, and the other prior art of record do not disclose “….determining that the speech recognition transcription contains a word that is missing from a local vocabulary; and, responsive to the determination, issuing a request to the server to send a description of the missing word; receiving the description of the missing word; and updating the local vocabulary with the description of the missing word” as recited by claim 3, or “…responsive to determining that the speech recognition transcription contains a word that is missing from a local vocabulary, issuing a request to the server to send a description of the missing word; performing local speech recognition on the spoken query, using a local vocabulary; and comparing the corresponding score from the remote server to a local speech recognition score, wherein the issuing of the request to the server to send the description of the missing word depends on the outcome of the comparison” as recited by claim 7.
Claims 4-6 depend on and further limit the allowable subject matter of intervening claim 3.
Claim 12 recites subject matter similar to that in claim 3, which is allowable for similar reasons.
Claim 13 depends on and further limits the allowable subject matter of intervening claim 12.
Claim 20 recites subject matter similar to that in claim 3, which is allowable for similar reasons.
Conclusion
The prior art made of record and not relied upon is considered pertinent to applicant's disclosure.
US 6092045 Stubley discloses speech recognition by dividing into blocks and comparing them to subsets representing models
US 7027987 Franz discloses a voice interface for a search engine
US 20030125869 Adams, JR. discloses creating a geographically limited vocabulary for a speech recognition system
Any inquiry concerning this communication or earlier communications from the examiner should be directed to Jesse Pullias whose telephone number is 571/270-5135. The examiner can normally be reached on M-F 8:00 AM - 4:30 PM. The examiner’s fax number is 571/270-6135.
If attempts to reach the examiner by telephone are unsuccessful, the examiner's supervisor, David Hudspeth can be reached on 571/272-7843.
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).
/Jesse S Pullias/
Primary Examiner, Art Unit 2657 10/07/26
