CTNF 18/820,282 CTNF 76255 DETAILED ACTION 07-03-aia AIA 15-10-aia The present application, filed on or after March 16, 2013, is being examined under the first inventor to file provisions of the AIA. Specification 07-29 AIA The disclosure is objected to because of the following informalities: On page 7, line 11, “decimates the ability” does not appear grammatical. On page 9, line 2, “may be used in when” is not grammatical but could be “may be used in” or “may be used when”. On page 10, line 15, “to increase the change” should be “to increase the chance”. On page 17, line 10, “system” should be “systems” . Appropriate correction is required. Claim Rejections - 35 USC § 112 07-30-02 AIA 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. 07-34-01 Claims 7 to 12 are rejected under 35 U.S.C. 112(b) or 35 U.S.C. 112 (pre-AIA), second paragraph, as being indefinite for failing to particularly point out and distinctly claim the subject matter which the inventor or a joint inventor (or for applications subject to pre-AIA 35 U.S.C. 112, the applicant), regards as the invention. Independent claims 7 and 12 set forth a limitation of “wherein the machine perception parameter differs from another parameter related to human perception”, where “another parameter related to human perception” is indefinite under 35 U.S.C. §112(b). Applicants’ claim language sets forth that “the machine perception parameter is related to a convergence time of less than 50 ms”, so it can be inferred that “the machine perception parameter” may relate to a filter parameter in echo cancellation, but a scope of the limitation is unclear for “another parameter related to human perception”. The Specification, Page 13, Lines 6 to 18, states that a machine perception parameter may be noise injection which differs from noise reduction for human perception, and a machine perception parameter may be echo scrambling, which differs from echo cancellation for human perception. Still, a scope of “another parameter related to human perception” is not a positive claim limitation because nothing in the operation of the device as set forth by the subsequent claim language actually requires this “another parameter”. That is, Applicants’ claim language requires “a machine perception parameter” to perform an operation of “performing machine perception”, but the claimed method and apparatus does not positively utilize “another parameter related to human perception”. The limitation of “another parameter related to human perception” is only set forth in contrast to “a machine perception parameter”, and is not well-defined. Conceivably, “a machine perception parameter” might differ from “another parameter related to human perception” simply because an audio signal is converted into an electrical signal for automatic speech recognition but an audio signal is simply heard directly as an audio waveform by a human ear. Claim Rejections - 35 USC § 103 07-20-aia AIA 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. 07-21-aia AIA Claim s 7 and 9 to 12 are rejected under 35 U.S.C. 103 as being unpatentable over Ayrapetian et al . (U.S. Patent No. 9,286,883) in view of Conne Skidanenko (“Ensuring Voice Quality with Adequate Tail Length”) . Concerning independent claims 7 and 12, Ayrapetian et al . discloses a method and device for acoustic echo cancellation with automatic speech recognition, comprising: “a processor; and a memory, wherein the processor is configured to control the apparatus to” – communication system 102 may be a computing device that includes one or more processors and a memory which may contain software applications executed by the processors (column 4, lines 11 to 17: Figure 1); “receiving an audio signal, wherein the audio signal corresponds to audio that has been captured by a device” – communication system 102 (“a device”) may include one or more microphones for accepting speech input (“an audio signal”) on which to perform speech recognition (column 4, lines 17 to 20: Figure 1); an audio content signal may be provided to a presentation device 104, and presentation device 104 may present the audio content at (C); audio content may include any audio or audiovisual content including a signal from a remote party to a phone conversation, a radio station, a music stream, an audio-visual stream, etc .; presented audio content may be received by microphone 122 of communication system 100 (column 5, lines 43 to 53: Figure 1); microphone 122 may receive a combination of presented audio content and utterance audio from a user (column 6, lines 14 to 18: Figure 2); AEC 120 may receive both a microphone input signal n and a reference signal x corresponding to an audio content signal; a user may be watching a movie and also speaking a command; a reference signal x includes the audio of the movie, and the utterance audio signal v includes the user’s utterance (column 8, lines 7 to 13: Figure 3: Step 310); “performing machine perception on the audio signal to generate a machine perception output . . . wherein the machine perception is performed based on speech recognition of the audio signal” – automatic speech recognition is used on signals from acoustic echo cancellers (AECs) that add random noise (column 3, lines 16 to 19); ASR module 124 can perform ASR on utterance audio and residual random noise (column 6, lines 35 to 36: Figure 2); AEC 120 can provide an output signal e to an ASR module 124; ASR module 124 can perform speech recognition on signal e (column 9, lines 5 to 26: Figure 3: Step 318); here, “machine perception” is simply equivalent to automatic speech recognition and “a machine perception output” is an output signal of automatic speech recognition; Compare Specification, Page 1, Lines 20 to 23, which appears to define machine perception as automatic speech recognition; “wherein the machine perception is performed based on a preconditioning and a machine perception parameter, wherein the machine perception parameter is related to a convergence [time of less than 50 ms]” – acoustic echo cancellation is performed using random noise; random noise may be introduced into a reference signal path and into a microphone signal path (“preconditioning”) (Abstract); maintaining a trained AEC that is converged may be particularly important for applications in which a device may be addressed through spoken commands without first activating speech recognition non-verbally, e.g. , by pressing a button; an AEC that is continuously converged (“related to a convergence”) can ensure that a truer signal is provided to an ASR system (column 3, lines 19 to 27); residual random noise may be desirable due to its ability to cover up errors and artifacts in acoustic echo cancellation (AEC) (column 3, lines 33 to 41); AEC module 120 performing process 300 can use random noise to remain converged (“related to a convergence”) (column 6, lines 53 to 57: Figure 3); random noise may be generated and added to a reference signal path and microphone signal path; noise generator 410 may generate random noise signal n, and random noise signal n may be combined with reference signal x (column 7, lines 20 to 34: Figure 3: Step 306); AEC 120 can cancel acoustic echo in signal w; an estimated echo from adaptive filter of AEC 120 may be subtracted from a combined microphone and modified random noise signal to produce an output given by Equations (5) and (6) (column 8, line 43 to column 9, line 4: Figure 3: Step 314); here, “preconditioning” is equivalent to adding random noise prior to echo cancellation and “a machine perception parameter” can be construed as a parameter of random noise injected into AEC 120; Compare Specification, Page 2, Lines 19 to 20, which describes “preconditioning” as adding noise to an audio signal; “wherein the machine perception parameter differs from another parameter related to human perception” – acoustic echo cancellers (AEC) typically use adaptive filters, but often leave residual noise that may interfere with downstream consumers of microphone input to automatic speech recognition systems; introducing random noise into an acoustic echo canceller reference signal path and microphone path enables an acoustic echo canceller to continue to have a signal on which to perform echo cancellation and filter adaptation even when a reference signal is not active or silent; by providing random noise in the reference signal path and in the microphone path, an adaptive filter of the acoustic echo canceller may remain trained and converged because it remains active (column 2, line 17 to column 3, line 15); broadly, “the machine perception parameter” can be construed as parameters of random noise injected into an acoustic echo canceller for automatic speech recognition, and “another parameter related to human perception” comprises parameters for an acoustic echo canceller without injected random noise. Concerning independent claims 7 and 12, Ayrapetian et al . discloses all of the limitations with the exception of a machine perception parameter related to a convergence “time of less than 50 ms”. Generally, Ayrapetian et al . discloses “preconditioning” an audio signal by adding random noise during echo cancellation so that coefficients of an adaptive filter remain converged for automatic speech recognition. Coefficients of an adaptive filter for acoustic echo cancellation with injected random noise for speech recognition can be construed as “machine perception parameters” and coefficients of an adaptive filter for acoustic echo cancellation without injected random noise can be construed as conventional “parameters related to human perception”. Compare Specification, Page 2, Lines 19 to 20, which describes ‘preconditioning’ the audio signal as adding noise to an audio signal to generate a preconditioned signal. The only question, then, is if “a convergence time of less than 50 ms” is an obvious convergence time given in the prior art. Concerning independent claims 7 and 12, Conne Skidanenko teaches an echo cancellation algorithm to ensure voice quality with adequate tail length. Generally, tail length is a distance measured in milliseconds between an echo canceller and a signal reflector. Specifically, Conne Skidanenko states that to accomplish superior voice quality, an echo canceller must have several characteristics including enough memory, converge within 50 ms, and provide a wide enough range that speech clipping is not noticeable. (Page 4) Conne Skidanenko , then, is maintained to teach that it is known to provide an echo canceller with parameters that enable it to converge in less than 50 ms. An objective is to accomplish superior voice quality. It would have been obvious to one having ordinary skill in the art to provide echo cancellation parameters having a convergence time of less than 50 ms as taught by Conne Skidanenko in echo cancellation for automatic speech recognition of Ayrapetian et al . for a purpose of accomplishing superior voice quality. Concerning claim 9, Ayrapetian et al . discloses that output may be used to perform speech recognition (Abstract); devices are configured to obtain audio data of user utterances that include both a loudspeaker and a microphone (column 1, lines 24 to 25); communication system 102 may include one or more microphones 122 for accepting speech input on which to perform speech recognition (column 4, lines 17 to 20: Figure 1); user 106 may make an utterance at (H), and microphone 122 may receive a reference signal and utterance audio from the user (column 6, lines 13 to 18: Figure 2); an audio utterance of a user is speech (“wherein the audio signal includes speech”). Concerning claim 10, Ayrapetian et al . discloses that a volume of noise added to the microphone can be dynamically determined based on the microphone output; if the sound coming in through the microphone is soft, then the volume of random noise added to the microphone may be lower than if the sound coming in through the microphone is very loud; using lower-levels of random noise can prevent or reduce the risk of obscuring desirable low-level sounds, and using higher levels of random noise can obscure undesirable high-level sounds (column 3, lines 51 to 61); a noise generator can determine whether a current production of random noise should be adjusted; the determination can be made with respect to the measurement of the microphone output obtained; it may be desirable to use a lower level of random noise so that important acoustic features are not obscured, and to use a higher level of random noise so that unimportant acoustic features of the microphone signal are obscured or dominated (column 10, lines 15 to 36: Figure 5). Adjusting an amount of random noise based on a microphone signal is “wherein the preconditioning id is a preconditioning based on the audio signal.” Concerning claim 11, Ayrapetian et al . discloses that a process 300 for echo cancellation can be embodied in a set of executable program instructions (column 6, lines 64 to 67: Figure 3); a process 500 may be embodied in a set of executable program instructions loaded into memory, e.g. , RAM, and executed by one or more processors (column 9, lines 52 to 61: Figure 5); a software module executed by a processor can reside in a non-transitory computer-readable storage medium, e.g. , RAM memory (column 11, lines 1 to 16) . 07-22-aia AIA Claim 8 is rejected under 35 U.S.C. 103 as being unpatentable over Ayrapetian et al . (U.S. Patent No. 9,286,883) in view of Conne Skidanenko (“Ensuring Voice Quality with Adequate Tail Length”) as applied to claim 1 above, and further in view of Cho et al. (U.S. Patent Publication 2009/0150146) . Ayrapetian et al . does not expressly disclose “wherein the audio signal is segmented into frames of 20 ms, and wherein the machine perception is performed on a 20 ms frame”. However, Cho et al. teaches a microphone array speech recognition system that extracts target speech spoken for speech recognition to obtain a high speech recognition rate even in a noise environment. (Abstract) Specifically, feature vectors for speech recognition are calculated from a 20 ms-interval frame every 10 ms of the N separated sound-source signals. (¶[0032]: Figure 2) Generally, it is maintained to be common and conventional to perform speech recognition on audio frames in a range of 10 ms to 40 ms as taught by Cho et al. An objective is to obtain a high speech recognition rate even in a noise environment. (Abstract) It would have been obvious to one having ordinary skill in the art to segment audio into frames of 20 ms to perform speech recognition as taught by Cho et al. in acoustic echo cancellation for automatic speech recognition of Ayrapetian et al . for a purpose of obtaining a high speech recognition rate even in a noise environment. Conclusion 07-96 The prior art made of record and not relied upon is considered pertinent to Applicants’ disclosure. Nosrati et al. is Applicants’ parent patent. Kleijn et al. discloses related prior art. Any inquiry concerning this communication or earlier communications from the examiner should be directed to MARTIN LERNER whose telephone number is (571) 272-7608. The examiner can normally be reached Monday-Thursday 8:30 AM-6:00 PM. If attempts to reach the examiner by telephone are unsuccessful, the examiner’s supervisor, Richemond Dorvil can be reached at (571) 272-7602. 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. /MARTIN LERNER/Primary Examiner Art Unit 2658 April 10, 2026 Application/Control Number: 18/820,282 Page 2 Art Unit: 2658 Application/Control Number: 18/820,282 Page 3 Art Unit: 2658 Application/Control Number: 18/820,282 Page 4 Art Unit: 2658 Application/Control Number: 18/820,282 Page 5 Art Unit: 2658 Application/Control Number: 18/820,282 Page 6 Art Unit: 2658 Application/Control Number: 18/820,282 Page 7 Art Unit: 2658 Application/Control Number: 18/820,282 Page 8 Art Unit: 2658 Application/Control Number: 18/820,282 Page 9 Art Unit: 2658 Application/Control Number: 18/820,282 Page 10 Art Unit: 2658
