Speech Signal Processing
Neurosoft began conducting research on the analysis and audio signals processing as early as at the beginning of the 1990’s. The direction of research related to the speech synthesis and analysis is consistent with the mission appointed by the Neurosoft founders – creating new solutions making work with computer easier.
The work on the speech synthesis has resulted in the development of the SynTalk program whose latest version 2.0 generates synthetic speech of high naturalness, using the mechanism of the diaphonic synthesis and the modeling of the prosody supported by grammatical analysis.
When it comes to speech recognition, so far we have created a prototype system to recognize short voice commands and a prototype system for indexing and phonetic search in the sound repositories.