A combined approach to the analysis of speech conversations in a contact center domain

TL;DR

This paper presents a comprehensive speech analytics approach for contact centers, including speech-to-text conversion, semantic tagging, and classification, demonstrating improved performance and interpretability in analyzing customer-agent conversations.

Contribution

It introduces an integrated methodology combining speech recognition, semantic tagging, and decision tree classification tailored for contact center data analysis.

Findings

Kaldi-based speech-to-text outperforms Google Cloud API in this context

Combining rule-based and machine learning tagging improves accuracy

J48S decision tree offers competitive performance with high interpretability

Abstract

The ever more accurate search for deep analysis in customer data is a really strong technological trend nowadays, quite appealing to both private and public companies. This is particularly true in the contact center domain, where speech analytics is an extremely powerful methodology for gaining insights from unstructured data, coming from customer and human agent conversations. In this work, we describe an experimentation with a speech analytics process for an Italian contact center, that deals with call recordings extracted from inbound or outbound flows. First, we illustrate in detail the development of an in-house speech-to-text solution, based on Kaldi framework, and evaluate its performance (and compare it to Google Cloud Speech API). Then, we evaluate and compare different approaches to the semantic tagging of call transcripts, ranging from classic regular expressions to machine learning models based on ngrams and logistic regression, and propose a combination of them, which is shown to provide a consistent benefit. Finally, a decision tree inducer, called J48S, is applied to the problem of tagging. Such an algorithm is natively capable of exploiting sequential data, such as texts, for classification purposes. The solution is compared with the other approaches and is shown to provide competitive classification performances, while generating highly interpretable models and reducing the complexity of the data preparation phase. The potential operational impact of the whole process is thoroughly examined.