Generating data features from speech and sentiment analytics for enhanced predictive routing

Patent metadata

Field	Value
Publication number	US-12499872-B2
Application number	US-202318393642-A
Country	US
Kind code	B2
Filing date	Dec 21, 2023
Priority date	Dec 21, 2023
Publication date	Dec 16, 2025
Grant date	Dec 16, 2025

Abstract

A method for processing data for training a predictive routing model. The method includes receiving interaction data from previous interactions that includes audio data capturing a conversation and transcript data of the conversation. The method continues by performing speech analytics by processing the audio data to determine scores for speech metrics that include a measure of how much the agent or customer speaks during the conversation. The method continues by performing sentiment analysis to determine scores associated with sentiment metrics, the sentiment metrics including a measure of a sentiment based on classifying utterances appearing in the transcript data as being positive or negative. The method continues by performing feature engineering to generate feature data and generating a training dataset therefrom. The method continues by applying a machine learning algorithm to the training dataset to train a predictive routing model.

First claim

That which is claimed:

1 . A method for processing data associated with interactions previously conducted by a contact center for training a predictive routing model, the method comprising:
receiving the data associated with the previously conducted interactions, wherein, when described in relation to a first interaction of the interactions, the received data comprises:
attribute data describing attributes of the first interaction, the attribute data including: interaction data describing one or more interaction attributes of the first interaction; customer data describing one or more customer attributes of a customer of the first interaction; agent data describing one or more agent attributes of an agent of the first interaction; and outcome data describing an outcome of the first interaction;
audio data that comprises audio of a recording of a conversation occurring between a customer and an agent during the first interaction;
transcript data comprising text of the conversation of the first interaction as generated via automatic speech recognition from the audio data of the first interaction;
performing speech analytics in relation to the received audio data of the interactions, wherein, when described in relation to the audio data of the first interaction, the speech analytics comprises processing the audio data of the first interaction to determine one or more scores associated with one or more speech metrics, the one or more speech metrics including a measure of an extent to which the agent or customer speaks during the conversation of the first interaction; performing sentiment analysis in relation to the transcript data of the interactions, wherein, when described in relation to the transcript data of the first interaction, the sentiment analysis comprises processing the transcript data of the first interaction to determine one or more scores associated with one or more sentiment metrics, the one or more sentiment metrics including a measure of a sentiment that is based on classifying utterances appearing in the transcript data of the first interaction as being positive or negative; performing feature engineering to generate feature data associated with the interactions, the generated feature data including:
attribute features derived from attributes data;
speech analytics features derived from the determined scores of the one or more speech metrics;
sentiment analysis features derived from the determined scores of the one or more sentiment metrics;
generating a training data set from the feature data, the training dataset including attribute features, speech analytics features, and sentiment analysis features; applying a machine learning algorithm to the training dataset to train a predictive routing model based on patterns identified within the training dataset whereby agent-customer pairings are correlated with a performance measure of contact center that is based, at least in part, on the outcomes described by the outcome data.

2 . The method of claim 1 , wherein the one or more speech metrics include at least one of a speaking metric; a silence metric; and an overlapping speech metric; and
wherein:
the speaking metric comprises a measure as to how much the agent speaks and how much the customer speaks during the conversation of the first interaction;
the silence metric comprises a measure as to how much silence occurs during the conversation of the first interaction; and
the overlapping speech metric comprises a measure as to how much during the conversation of the first interaction that the agent and the customer are both talking at the same time.

3 . The method of claim 2 , wherein the one or more speech metrics comprises each of the speaking metric, the silence metric; and the overlapping speech metric.

4 . The method of claim 3 , wherein the speaking metric comprises a customer speaking amount, which is an absolute length of time that the customer speaks during the conversation of the first interaction, and an agent speaking amount, which is an absolute length of time that the agent speaks during the conversation of the first interaction.

5 . The method of claim 3 , wherein the speaking metric comprises at least one of a customer speaking percentage and an agent speaking percentage,
wherein:
the customer speaking percentage comprises an absolute length of time that the customer speaks during the conversation of the first interaction divided by a conversation length, which is a total time of the conversation of the first interaction; and
the agent speaking percentage comprises an absolute length of time that the agent speaks during the conversation of the first interaction divided by the conversation length of the first interaction.

6 . The method of claim 3 , wherein the speaking metric comprises a ratio comparing a customer speaking amount against an agent speaking amount;
wherein:
the customer speaking amount comprises an absolute length of time that the customer speaks during the conversation of the first interaction; and
the agent speaking amount comprises an absolute length of time that the agent speaks during the conversation of the first interaction.

7 . The method of claim 3 , wherein the silence metric comprises a silence amount, the silence amount comprising an absolute length of time that silence occurs during the conversation of the first interaction.

8 . The method of claim 3 , wherein the silence metric comprises a silence percentage, the silence percentage comprising an absolute length of time that silence occurs during the conversation of the first interaction divided by a conversation length, which is a total time of the conversation of the first interaction.

9 . The method of claim 3 , wherein the silence metric comprises a ratio comparing a silence amount against at least one of a customer speaking amount and an agent speaking amount;
wherein:
the silence amount comprises an absolute length of time that silence occurs during the conversation of the first interaction;
the customer speaking amount comprises an absolute length of time that the customer speaks during the conversation of the first interaction; and
the agent speaking amount comprises an absolute length of time that the agent speaks during the conversation of the first interaction.

10 . The method of claim 3 , wherein the overlapping speech metric comprises an overlapping speech amount, the overlapping speech amount comprising an absolute length of time that overlapping speech occurs during the conversation of the first interaction.

11 . The method of claim 3 , wherein the overlapping speech metric comprises a overlapping speech percentage, the overlapping speech percentage comprising an absolute length of time that overlapping speech occurs during the conversation of the first interaction divided by a conversation length, which is a total time of the conversation of the first interaction.

12 . The method of claim 3 , wherein the overlapping speech metric comprises a ratio comparing an overlapping speech amount against at least one of a customer speaking amount and an agent speaking amount;
wherein:
the overlapping speech amount comprises an absolute length of time that overlapping speech occurs during the conversation of the first interaction;
the customer speaking amount comprises an absolute length of time that the customer speaks during the conversation of the first interaction; and
the agent speaking amount comprises an absolute length of time that the agent speaks during the conversation of the first interaction.

13 . The method of claim 3 , wherein the one or more sentiment metrics compris

Assignees

• Genesys Cloud Services Inc

Inventors

• Kalra Manan
• Munoz Emir
• Toth Greg
• Farrell David

Classifications

• G10L25/78

  Detection of presence or absence of voice signals (switching of direction of transmission by voice frequency in two-way loud-speaking telephone systems H04M9/10) · CPC title

• G10L15/28

  Constructional details of speech recognition systems · CPC title

• G10L15/16

  using artificial neural networks · CPC title

• G06F40/35

  Discourse or dialogue representation · CPC title

• H04M2203/401

  Performance feedback · CPC title