Ex Parte Hidaka et alDownload PDFPatent Trial and Appeal BoardJul 22, 201311397803 (P.T.A.B. Jul. 22, 2013) Copy Citation UNITED STATES PATENT AND TRADEMARK OFFICE _____________ BEFORE THE PATENT TRIAL AND APPEAL BOARD _____________ Ex parte KOTA HIDAKA, SHINYA NAKAJIMA, OSAMU MIZUNO, HIDETAKA KUWANO, and HARUHIKO KOJIMA _____________ Appeal 2011-004587 Application 11/397,803 Technology Center 2600 ______________ Before, KALYAN K. DESHPANDE, DAVID M. KOHUT, and MICHAEL J. STRAUSS, Administrative Patent Judges. KOHUT, Administrative Patent Judge. DECISION ON APPEAL Appeal 2011-004587 Application 11/397,803 2 This is a decision on appeal under 35 U.S.C. § 134(a) of the Final Rejection of claims 1-4, 6-13, 17, 19-27, 30, 31, and 34-37.1 An oral hearing was conducted on July 9, 2013. We have jurisdiction under 35 U.S.C. § 6(b). We affirm the Examiner’s rejection of these claims. INVENTION The invention is directed to a method, computer readable medium, and apparatus for processing speech in order to determine whether the speech was emphasized or not. Spec. 3-6. Claim 1 is representative of the invention and is reproduced below: 1. A speech processing method performed on a processor for deciding whether a portion of input speech is emphasized or not based on a set of speech parameters for each frame, comprising the steps of: (a) obtaining a code by quantizing a set of speech parameters for each of a plurality of frames in a portion of the input speech by using a codebook which stores, for each code, a speech parameter vector, a normal-state appearance probability and an emphasized-state appearance probability, each of said speech parameter vectors being composed of a set of speech parameters including at least two of a fundamental frequency, power and temporal variation of dynamic-measure and/or an inter-frame difference in at least one of those parameters, and obtaining from said codebook an emphasized-state appearance probability and a normal-state appearance probability corresponding to the code obtained for each frame in the portion; (b) calculating an emphasized-state likelihood with the processor of the portion by multiplying the emphasized-state appearance probabilities of the plurality of frames in the portion and calculating a normal-state likelihood of the portion by 1 Claims 5, 14-16, 18, 28, 29, 32, and 33 were previously cancelled. Appeal 2011-004587 Application 11/397,803 3 multiplying the normal-state appearance probabilities of the plurality of frames in the portion, respectively; and (c) deciding whether the portion of the input speech is emphasized or not based on said calculated emphasized-state likelihood and said normal-state likelihood, and outputting a decision result of said deciding, the decision result indicating whether the portion of the input speech is emphasized or not. REFERENCES Brown US 5,293,584 Mar. 8, 1994 Robinson US 5,999,899 Dec. 7, 1999 Thelen US 6,487,534 B1 Nov. 26, 2002 Chen, F.R.; Withgott, M., "The use of emphasis to automatically summarize a spoken discourse," Acoustics, Speech, and Signal Processing, 1992. ICASSP-92., 1992 IEEE International Conference on , vol.1, no., pp.229,232 vol.1, 23-26 Mar 1992, doi: 10.1109/ICASSP.1992.225930. (Hereinafter referred to as Chen). REJECTIONS AT ISSUE Claims 1-4, 6, 7, 10, 13, 17, 19-27, 30, 31, and 34-37 are rejected under 35 U.S.C. § 103(a) as being unpatentable over the combination of Chen, Robinson, and Applicants’ Admitted Prior Art (AAPA). Ans. 3-37. Claims 8 and 9 are rejected under 35 U.S.C. § 103(a) as being unpatentable over the combination of Chen, Robinson, Applicants’ Admitted Prior Art (AAPA), and Thelen. Ans. 38-41. Claims 11 and 12 are rejected under 35 U.S.C. § 103(a) as being unpatentable over the combination of Chen, Robinson, Applicants’ Admitted Prior Art (AAPA), Thelen, and Brown. Ans. 41-42. Appeal 2011-004587 Application 11/397,803 4 ISSUES Did the Examiner err in finding that the combination of Chen, Robinson, and AAPA teaches or suggests “a codebook which stores, for each code, a speech parameter vector, a normal-state appearance probability and an emphasized-state appearance probability, each of said speech parameter vectors being composed of a set of speech parameters,” as required by claim 1? ANALYSIS We select claim 1 as representative of claims 1-4, 6-13, 17, 19-27, 30, 31, and 34-37 as Appellants have not argued any of the other claims with particularity. 37 C.F.R. § 41.37(c)(1)(vii). Appellants argue that none of the references teach the disputed limitation. App. Br. 12. First, Appellants contend that Chen only teaches codebooks with a single speech parameter and not a vector with multiple speech parameters, as required by the claim. App. Br. 12. However, we note that the Examiner finds that Chen teaches multiple codebooks each with a corresponding parameter and finds that Robinson teaches a codebook wherein the speech parameter vector has a set of speech parameters. Ans. 42-44. Thus, the Examiner finds (Ans. 42-44), and we agree, that Robinson, not Chen, teaches a codebook wherein the vector contains multiple speech parameters. Second, Appellants contend that Robinson does not teach vectors that are composed of a set of different kinds of speech parameters. App. Br. 14; Reply Br. 5-6. Instead, Appellants contend that what the Examiner considers multiple speech parameters are merely the same parameter. App. Appeal 2011-004587 Application 11/397,803 5 Br. 14. We disagree. The Examiner finds that Robinson’s parameters include frequency and amplitude which are, by definition, different parameters that represent different information. Ans. 47. Additionally, the Examiner finds that Chen teaches F0 (fundamental frequency), energy, and delta-F0 (variation in pitch), as the different measured parameters. We agree with the Examiner (Ans. 47) and do not find Appellants’ argument to be persuasive. Third, Appellants argue that Robinson’s vectors do not contain appearance probabilities, as required by the claim. App. Br. 14. While we agree with Appellants’ assertion (App. Br. 14), the Examiner finds that Chen is used to teach appearance probabilities. Ans. 52. Thus, the Examiner finds that it is the combination of references that teaches normal-state and emphasized-state probabilities. Ans. 52. We agree with the Examiner. Lastly, Appellants argue that even if the combination of the references teaches all of the limitations, there is no rational reason to combine Robinson’s vector codebook with Chen and AAPA. App. Br. 15; Reply Br. 4. Specifically, Appellants argue that each of the references serve different purposes. Reply Br. 3-4. We disagree. Chen’s purpose is to summarize speech using codebooks. See Chen, pp. I-230-231. Robinson’s purpose is to encode and decode audio signals using codebooks, which are “most often used to model human speech signals.” See Robinson, col. 1, ll. 55-61 and col. 9, ll. 27-38. Thus, it would make sense to use the references together with the combination yielding no more than predictable results. KSR Int'l v. Teleflex Inc., 550 U.S. 398, 416-417 (2007). The Examiner also identifies the relevant portions of each of the references relied on throughout the Examiner’s Answer. See Ans. 3-7. To the extent that the Examiner relies on Appeal 2011-004587 Application 11/397,803 6 the knowledge of one of ordinary skill in the art to combine the teachings of the references, this practice is consistent with current case law. For example, in KSR at 418, the Supreme Court explains: Often, it will be necessary for a court to look to interrelated teachings of multiple patents; the effects of demands known to the design community or present in the marketplace; and the background knowledge possessed by a person having ordinary skill in the art, all in order to determine whether there was an apparent reason to combine the known elements in the fashion claimed by the patent at issue. To facilitate review, this analysis should be made explicit. See In re Kahn, 441 F.3d 977, 988 (C.A.Fed.2006) (“[R]ejections on obviousness grounds cannot be sustained by mere conclusory statements; instead, there must be some articulated reasoning with some rational underpinning to support the legal conclusion of obviousness”). As our precedents make clear, however, the analysis need not seek out precise teachings directed to the specific subject matter of the challenged claim, for a court can take account of the inferences and creative steps that a person of ordinary skill in the art would employ. In this case, the Examiner’s conclusions of obviousness are clearly articulated and are based on detailed factual findings that are supported by the references of record. See Ans. 3-7. Additionally, the Examiner finds that the combination would “provide greater efficiency.” Ans. 6 (citing Robinson, col. 11, ll. 45-58). We find the Examiner’s motivation to be reasonable. Thus, we agree with the Examiner. For all the reasons stated supra, we sustain the Examiner’s rejection of claims 1-4, 6-13, 17, 19-27, 30, 31, and 34-37. Appeal 2011-004587 Application 11/397,803 7 CONCLUSION The Examiner did not err in finding that the combination of Chen, Robinson, and AAPA teaches or suggests “a codebook which stores, for each code, a speech parameter vector, a normal-state appearance probability and an emphasized-state appearance probability, each of said speech parameter vectors being composed of a set of speech parameters,” as required by claim 1. SUMMARY The Examiner’s decision to reject claims 1-4, 6-13, 17, 19-27, 30, 31, and 34-37 is affirmed. No time period for taking any subsequent action in connection with this appeal may be extended under 37 C.F.R. § 1.136(a)(1)(iv). AFFIRMED msc Copy with citationCopy as parenthetical citation