# US 12,002,466 B2 · Systems and methods for avoiding inadvertently triggering a voice assistant > Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt - Number: US 12,002,466 B2 - Document ID: US12002466B2 - Status: Granted - Kind: B2 - Application number: 17/984,394 - Filed: 2022-11-10 - Published: 2023-08-10 - Granted: 2024-06-04 - Earliest claimed: 2020-04-29 - Assignees: Rovi Guides, Inc. - Applicants: Rovi Guides, Inc. - Inventors: Ankur Anil Aher; Jeffry Copps Robert Jose - Web page: https://patentgrep.com/patent/US12002466B2 ## CPC - G10L 15/22: Procedures used during a speech recognition process, e.g. man-machine dialogue - G10L 25/18: the extracted parameters being spectral information of each sub-band - G10L 25/60: for measuring the quality of voice signals - G10L 2015/223: {Execution procedure of a spoken command} - G10L: SPEECH ANALYSIS TECHNIQUES OR SPEECH SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING TECHNIQUES; SPEECH OR AUDIO CODING OR DECODING ## Abstract Systems and methods are provided herein for avoiding inadvertently trigging a voice assistant with audio played through a speaker. An audio signal is captured by sampling a microphone of the voice assistant at a sampling frequency that is higher than an expected finite sampling frequency of previously recorded audio played through the speaker to generate a voice data sample. A quality metric of the generated voice data sample is calculated by determining whether the generated voice data sample comprises artifacts resulting from previous compression or approximation by the expected finite sampling frequency. Based on the calculated quality metric, it is determined whether the captured audio signal is previously recorded audio played through the speaker. Responsive to the determination that the captured audio signal is previously recorded audio played through the speaker, the voice assistant refrains from being activated. ## Claims ### Claim 1 A method comprising: capturing an audio signal, by a microphone of a voice assistant, to generate a voice data sample; calculating a quality metric of the generated voice data sample by determining a number of artifacts resulting from previous compression or approximation by an expected finite sampling frequency in a period of the generated voice data sample and selecting, from a data structure having a plurality of quality metrics respectively corresponding to different numbers of artifacts in the period, the quality metric corresponding to the determined number of artifacts in the period; determining, based on comparing the calculated quality metric to a threshold, whether the captured audio signal is audio played through a speaker; in response to determining that the captured audio signal is the audio played through the speaker, refraining from activating the voice assistant; receiving user feedback to the refraining from activating the voice assistant; and adjusting the threshold based on the feedback. ### Claim 2 (depends on 1) The method of claim 1, wherein the capturing the audio signal comprises sampling the audio signal at a sampling frequency that is higher than the expected finite sampling frequency of the audio played through the speaker to generate a voice data sample. ### Claim 3 (depends on 1) The method of claim 1, wherein the capturing the audio signal by the microphone of the voice assistant is caused by detecting a keyword. ### Claim 4 (depends on 1) The method of claim 1, wherein the determining the number of artifacts in the voice data sample comprises: converting the voice data sample from the time domain to the frequency domain using a Fast Fourier Transform; and analyzing the voice data sample for noise components in a high frequency spectrum of the voice data sample in the frequency domain. ### Claim 5 (depends on 1) The method of claim 1, further comprising: capturing a subsequent audio signal, by the microphone of the voice assistant to generate a new voice data sample, wherein the quality metric of the new generated voice data sample is compared to the adjusted threshold. ### Claim 6 A system comprising: input/output circuitry configured to: capture an audio signal, by a microphone of a voice assistant, to generate a voice data sample; control circuitry configured to: calculate a quality metric of the generated voice data sample by determining a number of artifacts resulting from previous compression or approximation by an expected finite sampling frequency in a period of the generated voice data sample and selecting, from a data structure having a plurality of quality metrics respectively corresponding to different numbers of artifacts in the period, the quality metric corresponding to the determined number of artifacts in the period; determine, based on comparing the calculated quality metric to a threshold, whether the captured audio signal is audio played through a speaker; in response to determining that the captured audio signal is the audio played through the speaker, refrain from activating the voice assistants; receive user feedback to the refraining from activating the voice assistant; and adjust the threshold based on the feedback. ### Claim 7 (depends on 6) The system of claim 6, wherein the input/output circuitry is configured to capture the audio signal by: sampling the audio signal at a sampling frequency that is higher than the expected finite sampling frequency of the audio played through the speaker to generate a voice data sample. ### Claim 8 (depends on 6) The system of claim 6, wherein the input/output circuitry is configured to capture the audio signal by the microphone of the voice assistant by detecting a keyword. ### Claim 9 (depends on 6) The system of claim 6, wherein the control circuitry is configured to determine the number of artifacts in the voice data sample by: converting the voice data sample from the time domain to the frequency domain using a Fast Fourier Transform; and analyzing the voice data sample for noise components in a high frequency spectrum of the voice data sample in the frequency domain. ### Claim 10 (depends on 6) The system of claim 6, wherein the control circuitry is further configured to: capture a subsequent audio signal, by the microphone of the voice assistant to generate a new voice data sample, wherein the quality metric of the new generated voice data sample is compared to the adjusted threshold. ### Claim 11 A non-transitory computer readable medium comprising: instructions that when executed by input/output circuitry cause the input/output circuitry to: capture an audio signal, by a microphone of a voice assistant, to generate a voice data sample; instructions that when executed by control circuitry cause the control circuitry to: calculate a quality metric of the generated voice data sample by determining a number of artifacts resulting from previous compression or approximation by an expected finite sampling frequency in a period of the generated voice data sample and selecting, from a data structure having a plurality of quality metrics respectively corresponding to different numbers of artifacts in the period, the quality metric corresponding to the determined number of artifacts in the period; determine, based on comparing the calculated quality metric to a threshold, whether the captured audio signal is audio played through a speaker; in response to determining that the captured audio signal is the audio played through the speaker, refrain from activating the voice assistants; receive user feedback to the refraining from activating the voice assistant; and adjust the threshold based on the feedback. ### Claim 12 (depends on 11) The non-transitory computer readable medium of claim 11, wherein the instructions that cause the input/output circuitry to capture the audio signal cause the input/output circuitry to: sample the audio signal at a sampling frequency that is higher than the expected finite sampling frequency of audio played through the speaker to generate a voice data sample. ### Claim 13 (depends on 11) The non-transitory computer readable medium of claim 11, further comprising instructions that when executed by the input/output circuitry cause the input/output circuitry to: capture the audio signal by the microphone of the voice assistant by detecting a keyword. ### Claim 14 (depends on 11) The non-transitory computer readable medium of claim 11, wherein the instructions that cause the control circuitry to determine the number of artifacts in the voice data sample cause the control circuitry to: convert the voice data sample from the time domain to the frequency domain using a Fast Fourier Transform; and analyze the voice data sample for noise components in a high frequency spectrum of the voice data sample in the frequency domain. ### Claim 15 (depends on 11) The non-transitory computer readable medium of claim 11, further comprising instructions that when executed by control circuitry cause the control circuitry to: capture a subsequent audio signal, by the microphone of the voice assistant to generate a new voice data sample, wherein the quality metric of the new generated voice data sample is compared to the adjusted threshold.