NordenBladet — From today, the Riigikogu (Parliament of Estonia) is using a new system for preparing of verbatim reports of the sittings. The system is called Hans and it was developed by the Estonian IT company Finestmedia. The system is using the speech recognition technology created in Tallinn University of Technology (TalTech).

Administrative Director of the Chancellery of the Riigikogu Ahto Saks said that the new system performed the time-consuming work of stenographers. “The system records the debates taking place in the Riigikogu Session Hall as sound files, and the speech recognition writes it down. After that, editors come to help Hans and revise the text, and then it is published on the Riigikogu website,” Saks described the workflow.

Saks explained that it had taken two years from the idea to the introduction of the system, but it was not completely ready yet, and the development and improvements continued. “First, Hans will make the verbatim records of the Riigikogu sittings and Question Times. It is planned to use the system also in the work of the committees in the future,” Saks added.

Finestmedia’s Head of Division Lauri Esko said that creation of the verbatim reports system had been an exciting challenge for their company. The technological solutions provided by different parties could be inventively integrated to see how the synergy emerging from it brought about a major development.

“We created a solution that has a much wider potential than its today’s use,” Esko pointed out. “The systems containing speech recognition can be an efficient tool in the work of courts, prisons and the police, but also in health care, research and journalism, where writing down or making a detailed report of a conversation is necessary.” Esko added that the solution also had a great export potential. “We have already introduced Hans to delegations from several countries who wish to start writing the success story of their e-governance with its help,” he said.

Head of the Laboratory of Language Technology of Tallinn University of Technology Tanel Alumäe believes the new system shows how modern technology helps make routine work more effective.

“We have been actively working on speech recognition for 15 years, and today Estonian speech recognition is used by radiologists, researchers and journalists. Web-based speech recognition is freely available to everyone through ‘tekstiks.ee’ environment,” the researcher said. “In spite of the smallness of the Estonian language, the Estonian language technology is on a relatively high level.” Globally, speech recognition is widely used with larger languages.