# US 2021/0241258 A1 · Transaction management based on audio of a transaction > Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt - Number: US 2021/0241258 A1 - Document ID: US20210241258A1 - Status: Published application - Kind: A1 - Application number: 16/941,673 - Filed: 2020-07-29 - Published: 2021-08-05 - Earliest claimed: 2020-01-30 - Assignees: Capital One Service, LLC - Applicants: Capital One Service, LLC - Inventors: Michael Mossoba; Joshua Edwards; Abdelkader M'Hamed Benkreira - Web page: https://patentgrep.com/patent/US20210241258A1 ## CPC - G06Q 20/3272: {using an audio code} - G06N 20/00: Machine learning - G06Q 20/204: {comprising interface for record bearing medium or carrier for electronic funds transfer or payment credit} - G06Q 20/209: {Specified transaction journal output feature, e.g. printed receipt or voice output} - G06Q 20/322: {Aspects of commerce using mobile devices [M-devices]} - G06Q 20/3224: {Transactions dependent on location of M-devices} - G06Q 20/40145: {Biometric identity checks} - G06Q 20/42: Confirmation, e.g. check or permission by the legal debtor of payment - G10L 25/51: for comparison or discrimination - G10L 25/90: Pitch determination of speech signals - G06Q 20/3223: {Realising banking transactions through M-devices} - G06Q: INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR - G06N: COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS - G10L: SPEECH ANALYSIS TECHNIQUES OR SPEECH SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING TECHNIQUES; SPEECH OR AUDIO CODING OR DECODING ## Abstract An example transaction analysis platform is described herein. The transaction analysis platform may analyze an audio signal to determine whether a user is associated with a transaction. The transaction analysis platform may determine that a tone of the audio signal is a transaction tone associated with the transaction. The transaction analysis platform may process, based on detecting the transaction tone, the audio signal to identify a set of tones associated with the transaction. The transaction analysis platform may generate, based on a quantity of the set of tones, transaction information associated with the transaction, wherein the transaction information identifies a quantity of items associated with the transaction, wherein the quantity of items corresponds to the quantity of the set of tones. The transaction analysis platform may perform an action associated with the transaction information and a transaction account of the user. ## Claims ### Claim 1 A method, comprising: analyzing, by a device, an audio signal to detect a transaction tone associated with a transaction, wherein the transaction tone is associated with a type of transaction device at a location associated with a merchant; determining, by the device, an audio characteristic of the transaction tone; determining, by the device and based on the audio characteristic of the transaction tone, a probability that the transaction tone is associated with the transaction; determining, by the device, that the probability satisfies a threshold indicating that the transaction tone is associated with the transaction involving the merchant; prompting, by the device, a user to verify that the transaction tone is associated with the transaction involving the merchant; receiving, by the device, an indication of whether the transaction tone is verified; and causing, by the device and based on the indication of whether the transaction tone was verified, a machine learning model to be trained to use the transaction tone to detect transactions involving the merchant. ### Claim 2 (depends on 1) The method of claim 1, further comprising: processing, based on the audio characteristic of the transaction tone, the audio signal to identify a set of tones associated with the transaction and the audio characteristic; determining a quantity of the set of tones; and generating, based on the quantity of the set of tones, transaction information associated with the transaction, wherein the transaction information identifies a quantity of items associated with the transaction, wherein the quantity of items corresponds to the quantity of the set of tones. ### Claim 3 (depends on 2) The method of claim 2, wherein the transaction tone is determined to be one of the set of tones, and wherein the transaction tone corresponds to a first item that is registered in the transaction and remaining tones of the set of tones correspond to remaining items that are registered in the transaction. ### Claim 4 (depends on 2) The method of claim 2, wherein processing the audio signal includes filtering the audio signal to disregard other tones in the audio signal that do not have the audio characteristic. ### Claim 5 (depends on 1) The method of claim 1, wherein the machine learning model is trained based on one or more audio characteristics of one or more tones associated with a particular type of transaction device. ### Claim 6 (depends on 5) The method of claim 5, wherein historical data is used as input to the machine learning model. ### Claim 7 (depends on 1) The method of claim 1, further comprising: determining that the user is within a threshold distance of the location associated with the merchant; and causing, based on the user being within the threshold distance of the location, a microphone to be activated to generate the audio signal associated with the transaction involving the user and the merchant, wherein the microphone is activated when the user is determined to be within the threshold distance of the location and until the user is determined to not be within the threshold distance of the location. ### Claim 8 A device, comprising: one or more memories; and one or more processors communicatively coupled to the one or more memories, configured to: analyze an audio signal to detect a transaction tone associated with a transaction, wherein the transaction tone is associated with a type of transaction device at a location associated with a merchant; determine an audio characteristic of the transaction tone, wherein the audio characteristic is associated with one or more reference audio characteristics; process, based on the audio characteristic of the transaction tone, the audio signal to identify a set of tones associated with the transaction, wherein the set of tones are associated with the audio characteristic; determine, based on a quantity of the set of tones, a quantity of items associated with the transaction; generate, based on the quantity of items, transaction information associated with the transaction; and perform an action associated with the transaction information and a transaction account associated with a user. ### Claim 9 (depends on 8) The device of claim 8, wherein the transaction tone is determined to be one of the set of tones, and wherein the transaction tone corresponds to a first item that is registered in the transaction and remaining tones of the set of tones correspond to remaining items that are registered in the transaction. ### Claim 10 (depends on 8) The device of claim 8, wherein processing the audio signal includes filtering the audio signal to disregard other tones in the audio signal that do not have the audio characteristic. ### Claim 11 (depends on 8) The device of claim 8, wherein the one or more processors, when performing the action, are configured to: prompt the user to provide information associated an actual quantity of items associated with the transaction; receive feedback that indicates the actual quantity of items; and train a transaction analysis model based on the quantity of items, the actual quantity of transaction items, and the audio signal. ### Claim 12 (depends on 11) The device of claim 11, wherein the transaction analysis model is trained based on one or more audio characteristics of one or more tones associated with a particular type of transaction device. ### Claim 13 (depends on 8) The device of claim 8, wherein the one or more processors are further configured to: determine that the user is within a threshold distance of the location associated with the merchant; and cause, based on the user being within the threshold distance of the location, a microphone to be activated to generate the audio signal associated with the transaction involving the user and the merchant, wherein the microphone is activated when the user is determined to be within the threshold distance of the location and until the user is determined to not be within the threshold distance of the location. ### Claim 14 (depends on 8) The device of claim 8, wherein the one or more processors, when performing the action, are configured to: identify, based on a time or the location of the transaction, an entry of a transaction log that corresponds to the transaction, wherein the transaction log is associated with the transaction account; and append the transaction information to the entry to enable the entry to include the quantity of items involved in the transaction. ### Claim 15 A non-transitory computer-readable medium storing instructions, the instructions comprising: one or more instructions that, when executed by one or more processors, cause the one or more processors to: analyze an audio signal to detect a transaction tone associated with a transaction, wherein the transaction tone is associated with a type of transaction device at a location associated with a merchant; determine an audio characteristic of the transaction tone, wherein the audio characteristic is associated with one or more reference characteristics; determine, based on the audio characteristic of the transaction tone, a probability that the transaction tone is associated with the transaction; determine that the probability satisfies a threshold indicating that the transaction tone is associated with the transaction involving the merchant; generate, based on determining the probability satisfies the threshold, transaction information associated with the transaction; and perform an action associated with the transaction information and a transaction account of a user. ### Claim 16 (depends on 15) The non-transitory computer-readable medium of claim 15, wherein the one or more reference characteristics correspond to a different stage of a particular transaction. ### Claim 17 (depends on 15) The non-transitory computer-readable medium of claim 15, wherein the one or more instructions, when executed by the one or more processors, further cause the one or more processors to: process, based on the audio characteristic of the transaction tone, the audio signal to identify a set of tones associated with the transaction and the audio characteristic; and determine a quantity of the set of tones; and wherein the one or more instructions, that cause the one or more processors to generate the transaction information, cause the one or more processors to: determine the transaction information based on determining the quantity of the set of tones, wherein the transaction information identifies a quantity of items associated with the transaction, wherein the quantity of items corresponds to the quantity of the set of tones. ### Claim 18 (depends on 17) The non-transitory computer-readable medium of claim 17, wherein the transaction tone is determined to be one of the set of tones, and wherein the transaction tone corresponds to a first item that is registered in the transaction and remaining tones of the set of tones correspond to remaining items that are registered in the transaction. ### Claim 19 (depends on 17) The non-transitory computer-readable medium of claim 17, wherein the one or more instructions, that cause the one or more processors to perform the action, cause the one or more processors to: identify, based on a time or the location of the transaction, an entry of a transaction log that corresponds to the transaction, wherein the transaction log is associated with the transaction account; and append the transaction information to the entry to enable the entry to include the quantity of items involved in the transaction. ### Claim 20 (depends on 15) The non-transitory computer-readable medium of claim 15, wherein the one or more instructions, that cause the one or more processors to process the audio signal, cause the one or more processors to: filter the audio signal to disregard other tones in the audio signal that do not have the audio characteristic.