Semi-automated logging of contact center telephone calls

Abstract

Modern businesses use contact centers as a communication channel with users of their products and services. The largest factor in the expense of running a telephone contact center is the labor cost of its agents. IBM Research has built a new system, Contact-Center Agent Buddies (CAB), which is designed to help reduce the average handle time (AHT) for customer calls, thereby also reducing their cost. In this paper, we focus on the call logging subsystem, which helps agents reduce the time they spend documenting those calls. We built a Template CAB and a Call Logging CAB, using a pipeline consisting of audio capture of a telephone conversation, automatic speech recognition, text analysis, and log generation. We developed techniques for ASR text cleansing, including normalization of expressions and acronyms, domain terms, capitalization, and boundaries for sentences, paragraphs, and call segments. We found that simple heuristics suffice to generate high-quality logs from the normalized sentences. The pipeline yields a candidate call log which the agents can edit in less time than it takes them to generate call logs manually. Evaluation of the Call Logging CAB in an industrial contact center environment shows that it reduces the amount of time agents spend logging calls by at least 50% without compromising the quality of the resulting call documentation.

Past, present and future of contact centers: a literature review

“…Most current literature defines CCs as the physical station that allows an organization’s customers to contact them by different communication channels, such as telephone, touch-point telephone, fax, letter, e-mail, web, online live chat, and social networks. These communication channels have been studied in either a conceptual or an analytical way, for example: e-mail (Legros et al , 2013; Gupta et al , 2012; Li et al , 2012; Jolai et al , 2008; Malik et al , 2007; Nenkova and Bagga, 2003; Rodenstein, 2004), telephone (Murphy and Cerqua, 2012; Ali, 2011; Byrd et al , 2008; Millard and Hole, 2008; Balakrishnan and Munisamy, 2007; Koole, 2004; Armony and Maglaras, 2004; Lewis et al , 2002), instant message (Luo and Zhang, 2013; Sparks, 2012), live chat (Sparks, 2012; Mehrotra, 2003; Padmanabhan and Kummamuru, 2007; Steul, 2000), and social networks (Acharya et al , 2013; Burns and Friedman, 2012; Bordoloi et al , 2011; Schuster et al , 2011).…”

“…Several studies focus on other CC administration tasks, although the majority of research in relation to the analytical aspects of CCs falls under the call monitoring category. In this section, we review four of these applications and briefly explain them: Logging telephone calls: Byrd et al . (2008) proposed a model to log telephone calls (semi-automatic approach) to reduce CC costs by decreasing the time spent on call logging in comparison to a manual approach.…”

“…Logging telephone calls: Byrd et al . (2008) proposed a model to log telephone calls (semi-automatic approach) to reduce CC costs by decreasing the time spent on call logging in comparison to a manual approach.…”

Beyond Utterance Extraction: Summary Recombination for Speech Summarization

“…The specifics of call-centre transcript summarization are the focus of several papers. For instance in [7], call logs are generated by filling hand-written templates thanks to information extraction models, and these templates are complemented by unstructured data extraction. [8] in a set of different domains, and train an HMM that models both domain-specific and domain-wide topic sequences.…”

Improving hmm-based extractive summarization for multi-domain contact center dialogues

“…Although there has been an emergence of work that uses natural language processing techniques to improve the productivity of contact centers [1,2], little work has focused on the summarization of calls. One exception is [3], which uses manually created heuristic rules to extract key utterances, such as those related to caller intentions or the acceptance/loss of orders, and then uses those utterances to create call logs.…”