The Project VO.I.C.E FIRST

The Research and Development Project called VO.I.C.E FIRST, provides for the creation of an integrated software solution of Artificial Intelligence that will consist of a component for speech analysis and monitoring of social networks, and a prototype of Virtual Agents.

In particular, with regard to speech analysis and monitoring of social networks, Speech Recognition technologies will be exploited (process by which human oral language is recognized and subsequently processed through a computer) and Deep Learning (research field of automatic learning and ‘artificial intelligence that is based on different levels of representation, corresponding to hierarchies of characteristics of factors or concepts, where the high-level concepts are defined on the basis of the low-level ones) to translate the voice information of the contact center into textual data. These textual data will then be crossed with the information extracted from conversations available online in social networks through Artificial Intelligence algorithms. We will also take advantage of Machine Learning algorithms (a method of data analysis that automates the creation of an analytical model and allows computers to find hidden insights without being explicitly programmed to know where to look) and Artificial Intelligence to create Virtual Agents (programs who interpret natural language and, if properly trained, can dialogue with human interlocutors in order to provide information or perform certain operations) capable of managing conversations in natural language and via chat to provide support both to telephone operators and to end customers.

Objectives and Results

Artificial Intelligences are more and more evolved and refined. Several companies have started to use them for their Customer Care, through virtual assistants and software that can interact directly with customers, ensuring a quick and intuitive user experience.

The Research and Development Project called VO.I.C.E FIRST, therefore provides for the creation of an integrated Artificial Intelligence software solution consisting of:

  • Module 1: Speech Recognition technologies able to translate voice interactions into textual data;
  • Module 2:Social Media Brand Monitoring technologies for monitoring consumer interactions on the main Social Media;
  • Module 3: Chatbot technologies and Conversational Virtual Agents able to offer support to telephone operators and autonomously manage conversations via chat and in natural language with the customer, improving the user experience.

Speech Recognition

Through the use of Speech Recognin and Deep Learning technologies, it will be possible to achieve four primary objectives: reduce operating expenses, provide the best customer experience, increase revenues and reduce attraction (complaints).

The speech to text application will be able to track and understand the needs and desires of consumers, to interpret the level of satisfaction, to analyze the opinions on brand competitors, to measure the propensity to buy, to monitor loyalty phenomena churn, to identify new market trends and to seize new business opportunities (upsell / cross sell).

The predictive analyzes processed on Voice interactions will also allow for the identification of behavior patterns in voice recordings (customer, operator) which will allow to improve the efficiency in the day by day operational management of inbound / outbound telephone campaigns by improving customer engagement.

Social Media Brand Monitoring

The Web & Social Media Monitoring tool, through Artificial Intelligence algorithms, will allow you to obtain in real time the complete view of consumer behavior in relation to a brand.

The software solution will make it possible to cross-reference contact center data from speech to text with information extracted from conversations available online in social networks to thus obtain a complete view of consumer behavior in its relationship with a brand. The Social Media Monitoring project will be based on Artificial Intelligence algorithms and articulated in research, monitoring, focus, analysis and reporting.

 

 

 

 

 

Virtual and conversational agents (chatbot)

Finally, the integrated software solution provides for the use of Machine and Artificial Intelligence algorithms to create Virtual Agents capable of:

  • manage natural language conversations with end customers;
  • manage chat conversations with end customers; 
  • provide real-time support to operators engaged in managing telephone and chat interaction with end customers.

Finally, in their full functionality these solutions make it possible to create real intelligent dialogues with the service actors, based on the representation and processing of domain knowledge (Knowledge Based Conversational AI).

 

 

 

 

Impact

The integrated software solution that will allow you to obtain a complete view of consumer behavior in relation to a brand, thus improving the Customer Care service and sales performance.
This software solution, the result of the Research and Development project, will be used in daily practice with the aim of strategically repositioning the Teleperformance operating site in Puglia, transforming it from a telephone call center into an innovative omnichannel contact center, with an increase in the digital and analytical skills of committed human resources and greater employment protection in a constantly evolving market.

Project Code

KAQAWQ6

Duration

June 2020 – June 2023