Skip to Main Content
The performance of speech recognition systems is commonly degraded by either speech-related disabilities or by real-world factors such as the environmentpsilas noise level and reverberation. In this work, we propose a subvocal speech recognition system based on EMG signal for subvocal acquisition, Independent Component Analysis (ICA) for feature extraction and Neural Networks for classification. We have evaluated the systempsilas performance using a vowel phonemes database. The success rate was 93,99%.