HONEYWELL INTERNATIONAL INC.Download PDFPatent Trials and Appeals BoardMar 18, 20212019006591 (P.T.A.B. Mar. 18, 2021) Copy Citation UNITED STATES PATENT AND TRADEMARK OFFICE UNITED STATES DEPARTMENT OF COMMERCE United States Patent and Trademark Office Address: COMMISSIONER FOR PATENTS P.O. Box 1450 Alexandria, Virginia 22313-1450 www.uspto.gov APPLICATION NO. FILING DATE FIRST NAMED INVENTOR ATTORNEY DOCKET NO. CONFIRMATION NO. 15/089,970 04/04/2016 Robert E. De Mers H0053441 (002.2980) 2718 89955 7590 03/18/2021 HONEYWELL/LKGLOBAL Intellectual Property Services Group 300 S. Tryon Street Suite 600 Charlotte, NC 28202 EXAMINER CHAVEZ, RODRIGO A ART UNIT PAPER NUMBER 2658 NOTIFICATION DATE DELIVERY MODE 03/18/2021 ELECTRONIC Please find below and/or attached an Office communication concerning this application or proceeding. The time period for reply, if any, is set in the attached communication. Notice of the Office communication was sent electronically on above-indicated "Notification Date" to the following e-mail address(es): DL-PMT-SM-IP@Honeywell.com honeywell@lkglobal.com patentservices-us@honeywell.com PTOL-90A (Rev. 04/07) UNITED STATES PATENT AND TRADEMARK OFFICE BEFORE THE PATENT TRIAL AND APPEAL BOARD Ex parte ROBERT E. DE MERS and DAVID J. DOPILKA Appeal 2019-006591 Application 15/089,970 Technology Center 2600 Before ALLEN R. MACDONALD, MICHAEL J. STRAUSS, and JEREMY J. CURCURI, Administrative Patent Judges. CURCURI, Administrative Patent Judge. DECISION ON APPEAL STATEMENT OF THE CASE Pursuant to 35 U.S.C. § 134(a), Appellant1 appeals from the Examiner’s decision to reject claims 1–3, 5–7, 9–11, 13, 15–20, and 23–25. We have jurisdiction under 35 U.S.C. § 6(b). We AFFIRM IN PART. 1 We use the word Appellant to refer to “applicant” as defined in 37 C.F.R. § 1.42(a). Appellant identifies the real party in interest as Honeywell. Appeal Br. 2. Appeal 2019-006591 Application 15/089,970 2 CLAIMED SUBJECT MATTER The claims are directed to a “[s]ystem and method to distinguish sources in a multiple audio source environment.” Spec., Title. Claims 1 and 17, reproduced below, are illustrative of the claimed subject matter: 1. An audio system, comprising: a memory device comprising a speech recognition algorithm and audio tags; a processor coupled to the memory device, and to at least one of: a source of undesired audio signal and a source of desired audio signal, the source of the undesired audio signal and the source of the desired audio signal being separate, the processor configured to (i) determine when (a) it has access to the source of undesired audio signals; (ii) determine when (b) it has access to the source of desired audio signals; (iii) associate a respective audio tag with the source of desired audio signal based on (a) and (b), wherein the audio tag is overlaid on the respective audio signal, the audio tag being a carrier signal that is one of: audible, sub audible, and super audible; (iv) receive a collective audio signal comprising speech emitted from the source of undesired audio, referred to as an undesired audio signal, and speech emitted from the source of desired audio signal, referred to as a desired audio signal, the undesired audio signal and desired audio signal being concurrent; (v) based on the audio tag, eliminate the undesired audio signal from the collective audio signal; and (vi) responsive to eliminating the undesired audio signal, use the speech recognition algorithm to identify an intended Appeal 2019-006591 Application 15/089,970 3 command in the collective audio signal with the undesired audio signal eliminated. 17. An audio system, comprising: a memory device comprising speech recognition algorithms; a processor coupled to the memory device, the processor configured to (i) receive a collective audio signal comprising (a) an undesired audio signal comprising speech from a speaker device, (b) a desired audio signal comprising speech from a person, and wherein the speech from the person and the speech from the speaker is concurrent or overlapping, (ii) identify an audio tag in the collective audio signal, and (iii) eliminate the undesired audio signal from the collective audio signal prior to (a) performing speech recognition on the collective audio signal, and (b) identifying an intended command therein. Supp. Appeal Br. 2, 5 (Claims App.). REFERENCES The prior art relied upon by the Examiner is: Name Reference Date Lloyd US 2012/0022869 A1 Jan. 26, 2012 Humphrey US 8,713,593 B2 Apr. 29, 2014 Des Jardins US 2014/0270194 A1 Sept. 18, 2014 Appeal 2019-006591 Application 15/089,970 4 REJECTIONS Claims 17–19 are rejected under 35 U.S.C. § 102(a)(1) as anticipated by Lloyd. Final Act. 3–5. Claims 1–3, 5–7, 9–11, 13, 15, 16, and 23–25 are rejected under 35 U.S.C. § 103 as obvious over Lloyd and Des Jardins. Final Act. 5–16.2 Claim 20 is rejected under 35 U.S.C. § 103 as obvious over Lloyd and Humphrey. Final Act. 16–17. OPINION Obviousness Rejection of Claims 1–3, 5–7, 9–11, 13, 15, 16, and 23–25 over Lloyd and Des Jardins The Examiner finds Lloyd and Des Jardins teach all limitations of claim 1. Final Act. 5–8. In particular, the Examiner finds Lloyd teaches “(i) determine when (a) it has access to the source of undesired audio signals; (ii) determine when (b) it has access to the source of desired audio signals” as recited in claim 1. Final Act. 6–7 (citing Lloyd ¶¶ 22, 32, 44). Appellant presents the following principal arguments: Lloyd’s system 105 gets audio input from one source - the cell phone (See, FIG. 4). Applicants agree that the audio input from the cell phone provides Lloyd’s ASR engine 105 with a comingled audio signal of (i) desired audio (speech from the person) and (ii) geographic noise as the undesired audio. Applicants respectfully re-submit that, first, “geographic noise” is vague and likely to comprise many sources of sound waves, and second, that Lloyd does not teach or suggest that the ASR engine 105 has access to the source of the geographic noise. 2 Claims 23–25 are not included in the formal statement of the rejection of page 5 of the Final Action, but claims 23–25 are addressed on pages 15–16 of the Final Action. Appeal 2019-006591 Application 15/089,970 5 Appeal Br. 16; see also Reply Br. 3–5. In response, the Examiner explains “[t]his disclosure of acoustic models and noise models by Lloyd confirms that Lloyd is determining access using the stored models as similarly explained in p[aragraph] 0028 of the current specification.” Ans. 4. Lloyd’s system “has access to the source of undesired audio signals because it has access to the noise models,” and Lloyds system “has access to the source of desired audio signals because it has access to the acoustic models.” Ans. 5. Appellant’s arguments persuade us the Examiner erred in finding Lloyd teaches “(i) determine when (a) it has access to the source of undesired audio signals; (ii) determine when (b) it has access to the source of desired audio signals” as recited in claim 1. Appellant’s Specification discloses the following: In some embodiments, the system to distinguish sources 102 has access to at least one of: the source of undesired audio signal and the source of desired audio signal. In these embodiments, prior to operation, the system to distinguish sources 102 either (i) adds unique externally created audio tags to the one or more sources of the undesired audio signal or (ii) identifies a naturally occurring audio tag associated with the source of the desired audio signal. Spec. ¶ 28. Appellant’s Specification further discloses the following: In other embodiments, the system to distinguish sources 102 does not have access to any sources of the audio signal (neither the source(s) of undesired audio signal 104, or the source of desired audio signal 106); in such embodiments, the system to distinguish sources 102 audio processes the collective audio signal to identify naturally occurring audio tags therein, and then sorts or separates the audio signal in accordance with the identified naturally occurring audio tags. The identified naturally Appeal 2019-006591 Application 15/089,970 6 occurring audio tags are associated with audio sources and the associations are stored. Spec. ¶ 29. Thus, Appellant’s Specification distinguishes between embodiments where the system has access to sources and is thus able to process an individual audio signal from a source, and embodiments where the system does not have access to sources and is thus required to process the collective audio signal. Claim 1 requires “(i) determine when (a) it has access to the source of undesired audio signals; (ii) determine when (b) it has access to the source of desired audio signals.” Thus, claim 1 is directed to embodiments where the system has access to sources and is thus able to process an individual audio signal from a source. In contrast, Lloyd is directed to a system where the system does not have access to sources and is thus required to process the collective audio signal. See Lloyd ¶¶ 22 (“In addition to voices, the geotagged audio signals 107 to 109 may also include ambient sounds or environmental noises that occur (naturally or otherwise) at a particular location.” (emphasis added)), 32 (“Ambient noise or environmental sound portions of the audio signals may be stored by the ASR engine 105 for other purposes, for example to build geo-specific noise models.” (emphasis added)). Put another way, using stored acoustic models and noise models, as described in Lloyd, does not readily teach or suggest that the creation of such models had access to sources and was able to process an individual audio signal from a source, as required by our interpretation of claim 1. We, therefore, do not sustain the Examiner’s rejection of claim 1. We also do not sustain the Examiner’s rejection of claims 2, 3, 5–7, 9, 10, and 23–25, which depend from claim 1. Appeal 2019-006591 Application 15/089,970 7 Independent claim 11 recites “determining when (a) it has access to the source of undesired audio signals; determining when (b) it has access to the source of desired audio signals.” We, therefore, do not sustain the Examiner’s rejection of claim 11 for the same reasons discussed above when addressing claim 1. We also do not sustain the Examiner’s rejection of claims 13, 15, and 16, which depend from claim 11. Anticipation Rejection of Claims 17–19 by Lloyd The Examiner finds Lloyd discloses all limitations of claim 17. Final Act. 3–4. In particular, the Examiner finds Lloyd discloses “(ii) identify an audio tag in the collective audio signal . . . , and (iii) eliminate the undesired audio signal from the collective audio signal prior to (a) performing speech recognition on the collective audio signal,” as recited in claim 17. Final Act. 4 (citing Lloyd ¶¶ 63, 77); see also Ans. 6–7 (citing Lloyd ¶¶ 31–42). Appellant presents the following principal argument: Lloyd does not disclose “identifying an audio tag in the collective audio signal and eliminating the undesired audio signal from the collective audio signal prior to performing speech recognition.” Appeal Br. 20; see also Reply Br. 3–5. Appellant’s arguments do not persuade us of any reversible error in the Examiner’s findings. In contrast to claims 1 and 11, claim 17 does not require access to sources. Using stored acoustic models and noise models, as described in Lloyd, discloses the disputed limitations of claim 17. Appeal 2019-006591 Application 15/089,970 8 Lloyd discloses In general, the one or more acoustic models 111, along with noise models, language models, and/or other models, are applied to the audio signal 112 to translate or transcribe the spoken utterance 113 into one or more textual, candidate transcriptions 115, and to generate speech recognition confidence scores to the candidate transcriptions 115. In particular, the one or more acoustic models 111 include statistical representations of the sounds that make up each word of a particular language, and the noise models are used for noise suppression or noise compensation. Both models enhance the intelligibility of the spoken utterance 113 to the ASR engine 105. Lloyd ¶ 44 (emphasis added). Lloyd further discloses the speech recognition module 124 of the ASR engine 105 performs speech recognition on the audio signal 112 using the one or more geo-specific acoustic models 111 for the geographic location associated with the audio signal 112. When the audio signal 112 includes metadata that describes a device type of the mobile device 102, the ASR engine 105 may apply an acoustic model that is specific to both the geographic location associated with the audio signal, and to the device type of the mobile device 102. The speech recognition module 124 may generate one or more candidate transcriptions 115 that match the utterance encoded in the audio signal 112, and speech recognition confidence values for the candidate transcriptions. Lloyd ¶ 63 (emphasis added); see also Lloyd ¶ 77 (disclosing a geotagged signal). Thus, Lloyd discloses “(ii) identify an audio tag in the collective audio signal” (see Lloyd ¶¶ 63, 77), “and (iii) eliminate the undesired audio signal from the collective audio signal prior to (a) performing speech recognition on the collective audio signal” (see Lloyd ¶ 44), as recited in claim 17. Appeal 2019-006591 Application 15/089,970 9 We, therefore, sustain the Examiner’s rejection of claim 17. We also sustain the Examiner’s rejection of claims 18 and 19, which are not separately argued with particularity. Obviousness Rejection of Claim 20 over Lloyd and Humphrey Appellant argues “Humphrey does not remedy the deficiencies of Lloyd.” Appeal Br. 21. For reasons explained above, we determine Lloyd is not deficient. We, therefore, sustain the Examiner’s rejection of claim 20. CONCLUSION The Examiner’s decision to reject claims 1–3, 5–7, 9–11, 13, 15, 16, and 23–25 is reversed. The Examiner’s decision to reject claims 17–20 is affirmed. DECISION SUMMARY In summary: Claim(s) Rejected 35 U.S.C. § Reference(s)/Basis Affirmed Reversed 17–19 102(a)(1) Lloyd 17–19 1–3, 5–7, 9– 11, 13, 15, 16, 23–25 103 Lloyd, Des Jardins 1–3, 5–7, 9– 11, 13, 15, 16, 23–25 20 103 Lloyd, Humphrey 20 Overall Outcome 17–20 1–3, 5–7, 9– 11, 13, 15, 16, 23–25 Appeal 2019-006591 Application 15/089,970 10 TIME PERIOD FOR RESPONSE No time period for taking any subsequent action in connection with this appeal may be extended under 37 C.F.R. § 1.136(a). See 37 C.F.R. § 1.136(a)(1)(iv). AFFIRMED IN PART Copy with citationCopy as parenthetical citation