Skip to main content

Case studies

Voice analytics for quality

In an increasingly competitive market the quality, speed, accuracy, compliance and controllability of customer service are becoming extremely complex. With the solution developed for OTP Bank, the financial institution has improved the quality control of its customer service.

The OTP Group is Hungary's largest financial services provider and a leading player in the Central and Eastern European region. It provides high-quality solutions to the financial needs of 13 million customers in 9 countries, through more than 1300 branches and state-of-the-art electronic channels.


In a competitive market, the quality, speed, accuracy, compliance and controllability of customer service are extremely complex:

  • Only 1% of calls are evaluated, and more than 600 hours of recorded telephone conversations a day are unprocessed, unverified. There is no complete picture of how the call centre works.
  • The human evaluation criteria are not consistent and have a lot of subjective elements, making them difficult to apply for objective performance evaluation. Service levels have to be raised without increasing costs.


Successfully monitor the efficiency of call centre and operators by processing 100% of the incoming calls.


Modular implementation of a system which combines different NLP techniques (text mining, speech recognition), is able to transcribe and analyse the large amount of audio data that is continuously generated in the company.


  • Consultation, definition of success criteria.
  • 6-week PILOT period.


  • Ensuring parallel operation during the implementation of the CLEMVOICE system.
  • To evaluate the conversations as accurately as possible, the different call types were identified, evaluated and modelled according to different criteria.
  • 10+ quality assurance aspects were checked in each call.
  • Negative index model to check the conversation’s mood.
  • Automatic measurement of important call center KPIs in every call such as FCR and SQM.