Acquisition and processing of verbal information in military systems for prevention and suppression of crime and terrorism
- Professor Grazyna Demenko.
The institutions performing the project:
- Poznan Supercomputing and networking Center – Project Leader;
- Foundation of the Adam Mickiewicz University in Poznan;
- Future Voice System Sp. z o.o.
The objective of the project is to develop a system for recognizing verbal and non-verbal contents of speech that characterize the unique properties of the speaker’s voice in order to optimize speech recognition systems and systems for voice control of automatic equipment. The results of the project are intended to be used by the Polish Armed Forces, the Police, and other uniformed services.
The system will be have a modular structure. The following modules are to be developed:
- A speech recognition and automatic voice control module:
- automatic transcription of speech;
- issuing commands to automatic equipment in dictation mode;
- issuing commands to automatic equipment in command mode.
- A speaker recognition module:
- characteristics of the speaker’s voice;
- identification and verification of the speaker.
The solution to be developed will be used, among others, as a tool for automatic search in large voice recordings databases in order to find specific speakers (to be used by both law enforcement agencies in their operations and during court proceedings in order to achieve a more effective automatic recording of court hearings).
- Supporting speech synthesis module.This tool will enable reading of a text identified by the Automatic Speech Recognition System.
Project financed by The National Centre for Research and Development