Rationale
Speech and language technology applications are becoming increasingly important in everyday life, leading to the development of well-known applications, including computer assisted language learning and digital assistants (Siri and Cortana, for instance). In order to develop the complex algorithms that drive these applications, a thorough knowledge of human hearing, speech perception, linguistics and psychoacoustics is required.
Course Description
This course introduces the processing of speech signals by computer. Students will investigate the way speech is represented in different domains. Computational approaches to speech parameter extraction for speech technology applications are studied. The approach taken is cross-disciplinary, examining speech production and perception from the perspectives of linguistics, phonetics, hearing science and psychoacoustics. It also introduces and applies digital signal processing techniques in the extraction of important parameters of speech.
Speaking; Hearing; Sounds and symbols; Articulatory and acoustic phonetics; Phonology; Prosody; Speech spectra; Sampling; Fourier Transform; Linear filters; Linear prediction; Cepstral analysis
• Two equally weighted programming projects (50%)
• Two equally weighted hour-long In-course Tests (20%)
• One two-hour final written examination (30%)
Students must pass both coursework and exam components, separately.
On successful completion of the course, students should be able to:
Our 7 faculties and 12 professional schools offer more than 200 programmes to some 18,000 graduate, undergraduate and continuing studies students.
The UWI, Mona ranks first in Jamaica among accredited tertiary-level programmes. In 2012, the University was again one of Jamaica’s Top 100 Employers.
