Advancing Electromyographic Continuous Speech Recognition: Signal Preprocessing and Modeling
Institution: | Universität Karlsruhe |
---|---|
Department: | |
Year: | 2014 |
Keywords: | Electromyography, EMG-based Speech Recognition, Silent Speech interfaces |
Record ID: | 1098914 |
Full text PDF: | http://digbib.ubka.uni-karlsruhe.de/volltexte/documents/3087444 |
Speech is the natural medium of human communication, but audible speech can be overheard by bystanders and excludes speech-disabled people. This work presents a speech recognizer based on surface electromyography, where electric potentials of the facial muscles are captured by surface electrodes, allowing speech to be processed nonacoustically. A system which was state-of-the-art at the beginning of this thesis is substantially improved in terms of accuracy, flexibility, and robustness.