Smarter call centre automation for public administrations

Developed by the IST-funded HOPS project, the platform uses a variety of technologies to enable people to talk to a computer over the phone as if they are talking with a human call centre worker.

For public administrations, which are under increasing pressure to improve the quality of the services they provide to citizens while reducing costs, the benefits are significant in terms of efficiency gains and potential cost savings. The advantages, says Joan Batlle, the project coordinator at Barcelona city hall, will also be evident for citizens, who will find they have access to interactive information and services around the clock. Most importantly, the natural language dialogue made possible by the platform should overcome people’s general dislike of talking to automated call centre systems.

“Evidently people don’t like having to sit and listen to a machine, they feel it is a waste of time especially when it may not understand them,” explains the coordinator. “The problem has been that most of the systems used to automate call centres rely on asking the caller to pick from a list of options and then offering a response from a pre-programmed list that may or may not give the user the information they require.”

The HOPS partners, among them the three public administrations of Barcelona, Camden and Turin as well as research institutes and technology providers, have instead of relying on scripted responses, combined four different technologies to create an advanced voice-enabled front-end platform for public administrations.

“The idea is that people can speak naturally and receive a natural response. In that way it is the citizen who is in charge of the conversation not the computer,” Batlle notes. “I want to be able to ring up and say ‘What concerts are on tonight?’ without having to listen to a list of possible options first. I should be understood no matter how I phrase the question.”

The HOPS project has managed to make human-machine dialogue more natural and fluid by merging voice technologies such as Automatic Speech Recognition (ASR) and Text to Speech (TTS) with natural language processing technologies to understand, interpret and respond to callers. These components are then tied into a data management system incorporating Semantic Web technology for finding and extracting the information sought by users.

“The platform itself is designed to be highly flexible so it can be used in any public administration call centre to provide any service or information. The only thing that really has to be changed depending on where it is deployed is the vocabulary. That could mean different languages or a different lexicon depending on whether it is used to deal with car registrations or cultural events,” Batlle notes.

Preliminary trials carried out in Barcelona, Camden and Turin proved the flexibility and functionality of the system which was able to provide responses in a range of languages to callers asking about two different types of services.

“One scenario was for callers interested in finding out about cultural events, the other was to schedule a service provided by the council for collecting unwanted furniture and other large items,” the coordinator says. “In both cases the call centre workers from the councils made the calls and the system functioned well.”

The project is due to begin testing a second prototype in the autumn ahead of tests with a third and final version of the platform that will also serve to gauge citizens’ reactions.

Batlle says that all three town halls involved in the project are planning to employ the finished version of the platform. In Turin, the city hall is planning to use it to handle calls while retaining human operators as a backup, while in Camden the system will be employed initially to carry out quality of service surveys.

“In Barcelona we will start small, using it to replace the recorded messages we have overnight and at weekends although over the coming years we expect to use it more extensively, as too do the partners in Camden and Turin,” he says.