# US 12,190,863 B2 · System and method for automated processing of digitized speech using machine learning > Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt - Number: US 12,190,863 B2 - Document ID: US12190863B2 - Status: Granted - Kind: B2 - Application number: 17/750,573 - Filed: 2022-05-23 - Published: 2023-12-14 - Granted: 2025-01-07 - Earliest claimed: 2022-05-23 - Assignees: Conduent Business Services, LLC - Applicants: Conduent Business Services, LLC - Inventors: Dennis F. Quebe; Jian Feng; Ambrish Gupta; Ashwin Subramanyam - Web page: https://patentgrep.com/patent/US12190863B2 ## CPC - G10L 15/01: Assessment or evaluation of speech recognition systems - G10L 15/063: {Training} - G10L 15/06: Creation of reference templates; Training of speech recognition systems, e.g. adaptation to the characteristics of the speaker's voice (G10L15/14 takes precedence) - G10L 15/08: Speech classification or search - G10L 15/32: Multiple recognisers used in sequence or in parallel; Score combination systems therefor, e.g. voting systems - G10L: SPEECH ANALYSIS TECHNIQUES OR SPEECH SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING TECHNIQUES; SPEECH OR AUDIO CODING OR DECODING ## Abstract Automated systems and methods are provided for processing speech, comprising obtaining a trained machine learning model that has been trained using a cumulative historical data structure corresponding to at least one digitally-encoded speech representation for a plurality of telecommunications interactions conducted by a plurality of agent-side participants, which includes a first data corresponding to a score variable and a second data corresponding to a plurality of driver variables; applying the trained machine learning model: to a subset of data in the cumulative historical data structure that corresponds to a first agent-side participant of the plurality of agent-side participants, to generate a performance classification score and/or a performance direction classification score, to identify an intervention-target agent-side participant from among the plurality of agent-side participants, and to the cumulative historical data structure to identify an intervention training plan; and conducting at least one training session according to the intervention training plan. ## Claims ### Claim 1 A computer-implemented method for processing speech, the method comprising: obtaining a trained machine learning model, wherein the machine learning model has been trained using a cumulative historical data structure corresponding to at least one digitally-encoded speech representation for a plurality of telecommunications interactions conducted by a plurality of agent-side participants, wherein the cumulative historical data structure includes a first data corresponding to a score variable and a second data corresponding to a plurality of driver variables; determining a driver variable state as a function of the cumulative historical data structure; applying the trained machine learning model to a subset of data in the cumulative historical data structure that corresponds to a first agent-side participant of the plurality of agent-side participants, to generate at least one of a performance classification score or a performance direction classification score; applying the trained machine learning model to identify an intervention-target agent-side participant from among the plurality of agent-side participants; applying the trained machine learning model to the cumulative historical data structure to identify an intervention training plan, wherein identifying the intervention training plan comprises identifying one or more intervention variables for the intervention-target agent-side participant based on the driver variable state; and conducting at least one training session for the intervention-target agent-side participant according to the intervention training plan. ### Claim 2 (depends on 1) The method according to claim 1, wherein the performance classification score indicates a performance of the first agent-side participant relative to a plurality of second agent-side participants of the plurality of agent-side participants, and the performance direction classification score indicates a current performance of the first agent-side participant relative to a previous performance of the first agent-side participant. ### Claim 3 (depends on 1) The method according to claim 1, wherein identifying the intervention-target agent-side participant includes: applying the trained machine learning model for each of the plurality of agent-side participants, to calculate a potential lift value using potential values for at least one of the plurality of driver variables; and based on the potential lift value, classifying each of the plurality of agent-side participants as the intervention-target agent-side participant or a non-target agent-side participant. ### Claim 4 (depends on 1) The method according to claim 1, wherein identifying the intervention training plan includes: identifying an intervention-target two-dimensional data structure from among the cumulative historical data structure, wherein the intervention-target two-dimensional data structure corresponds to the intervention-target agent-side participant; and based on data corresponding to a plurality of intervention variables included in the intervention-target two-dimensional data structure, selecting the intervention training plan from among a plurality of candidate training plans. ### Claim 5 (depends on 1) The method according to claim 1, further comprising: generating the cumulative historical data structure, including: obtaining a plurality of agent values corresponding to the plurality of agent-side participants, wherein the plurality of agent values include a plurality of driver values and a score value, and generating a plurality of two-dimensional data structures each corresponding to one of the plurality of agent-side participants, wherein a respective row of the two-dimensional data structure includes a plurality of entries respectively corresponding to a cumulative average value of one of the plurality of agent values; and training the machine learning model by applying an algorithm to the plurality of two-dimensional data structures, wherein the algorithm is configured to treat the score value as a dependent variable and to treat each of the plurality of driver values as an independent variable. ### Claim 6 (depends on 1) The method according to claim 1, wherein the cumulative historical data structure includes a plurality of two-dimensional data structures, wherein respective ones of the plurality of two-dimensional data structures correspond to respective ones of the plurality of agent-side participants. ### Claim 7 (depends on 6) The method according to claim 6, wherein: a respective two-dimensional data structure includes N rows, where Nis a total number of telecommunication interactions conducted by a corresponding agent-side participant; a j-th row of the respective two-dimensional data structure corresponds to a cumulative average value for the first j telecommunication interactions conducted by the corresponding agent-side participant; and the respective two-dimensional data structure includes a score column containing values of the score variable, and a plurality of driver columns containing values of the plurality of driver variables. ### Claim 8 (depends on 1) The method according to claim 1, wherein the at least one digitally-encoded speech representation includes digitally-encoded data corresponding to at least one of a recording or a transcript of a telecommunications interaction. ### Claim 9 (depends on 1) The method according to claim 1, wherein identifying the intervention training plan further comprises identifying non-intervention variables. ### Claim 10 (depends on 1) The method according to claim 1, further comprising: obtaining a trained classification model, wherein the classification model has been trained using features extracted from the at least one digitally-encoded speech representation; and using the trained classification model, identifying the plurality of driver variables from among a plurality of variables included in the digitally-encoded speech representation. ### Claim 11 A computing system for processing speech, the system comprising: at least one electronic processor; and a non-transitory computer readable medium storing instructions that, when executed by the at least one electronic processor, cause the electronic processor to perform operations comprising: obtaining a trained machine learning model, wherein the machine learning model has been trained using a cumulative historical data structure corresponding to at least one digitally-encoded speech representation for a plurality of telecommunications interactions conducted by a plurality of agent-side participants, wherein the cumulative historical data structure includes a first data corresponding to a score variable and a second data corresponding to a plurality of driver variables, determining a driver variable state as a function of the cumulative historical data structure, applying the trained machine learning model to a subset of data in the cumulative historical data structure that corresponds to a first agent-side participant of the plurality of agent-side participants, to generate at least one of a performance classification score or a performance direction classification score, applying the trained machine learning model to identify an intervention-target agent-side participant from among the plurality of agent-side participants, applying the trained machine learning model to the cumulative historical data structure to identify an intervention training plan, wherein identifying the intervention training plan comprises identifying one or more intervention variables for the intervention-target agent-side participant based on the driver variable state, and facilitating at least one training session for the intervention-target agent-side participant according to the intervention training plan. ### Claim 12 (depends on 11) The system according to claim 11, wherein the performance classification score indicates a performance of the first agent-side participant relative to a plurality of second agent-side participants of the plurality of agent-side participants, and the performance direction classification score indicates a current performance of the first agent-side participant relative to a previous performance of the first agent-side participant. ### Claim 13 (depends on 11) The system according to claim 11, wherein identifying the intervention-target agent-side participant includes: applying the trained machine learning model for each of the plurality of agent-side participants, to calculate a potential lift value using potential values for at least one of the plurality of driver variables; and based on the potential lift value, classifying each of the plurality of agent-side participants as the intervention-target agent-side participant or a non-target agent-side participant. ### Claim 14 (depends on 11) The system according to claim 11, wherein identifying the intervention training plan includes: identifying an intervention-target two-dimensional data structure from among the cumulative historical data structure, wherein the intervention-target two-dimensional data structure corresponds to the intervention-target agent-side participant; and based on data corresponding to a plurality of intervention variables included in the intervention-target two-dimensional data structure, selecting the intervention training plan from among a plurality of candidate training plans. ### Claim 15 (depends on 11) The system according to claim 11, the operations further comprising: generating the cumulative historical data structure, including: obtaining a plurality of agent values corresponding to the plurality of agent-side participants, wherein the plurality of agent values include a plurality of driver values and a score value, and generating a plurality of two-dimensional data structures each corresponding to one of the plurality of agent-side participants, wherein a respective row of the two-dimensional data structure includes a plurality of entries respectively corresponding to a cumulative average value of one of the plurality of agent values; and training the machine learning model by applying an algorithm to the plurality of two-dimensional data structures, wherein the algorithm is configured to treat the score value as a dependent variable and to treat each of the plurality of driver values as an independent variable. ### Claim 16 (depends on 11) The system according to claim 11, wherein the cumulative historical data structure includes a plurality of two-dimensional data structures, wherein respective ones of the plurality of two-dimensional data structures correspond to respective ones of the plurality of agent-side participants. ### Claim 17 (depends on 16) The system according to claim 16, wherein: a respective two-dimensional data structure includes N rows, where Nis a total number of telecommunication interactions conducted by a corresponding agent-side participant; a j-th row of the respective two-dimensional data structure corresponds to a cumulative average value for the first j telecommunication interactions conducted by the corresponding agent-side participant; and the respective two-dimensional data structure includes a score column containing values of the score variable, and a plurality of driver columns containing values of the plurality of driver variables. ### Claim 18 (depends on 11) The system according to claim 11, wherein the at least one digitally-encoded speech representation includes digitally-encoded data corresponding to at least one of a recording or a transcript of a telecommunications interaction. ### Claim 19 (depends on 11) The system according to claim 11, wherein identifying the intervention training plan further comprises identifying non-intervention variables. ### Claim 20 (depends on 11) The system according to claim 11, the operations further comprising: obtaining a trained classification model, wherein the classification model has been trained using features extracted from the at least one digitally-encoded speech representation; and using the trained classification model, identifying the plurality of driver variables from among a plurality of variables included in the digitally-encoded speech representation.