Semester.ly

Johns Hopkins University | EN.520.680

Speech and Auditory Processing by Humans and Machines

3.0

credits

Average Course Rating

(4.34)

The course relevant to building advanced systems for information extraction from speech and auditory signals. It introduces some relevant historical efforts for information processing of speech and audio signals and basic concepts of human auditory perception and human production and perception of speech. The main goal of the course is in implementation of relevant knowledge of human speech information processing in engineering systems for information extraction from speech signals, emphasizing power of the modern data-guided machine learning techniques. Basic knowledge of signal processing is assumed and the previous completion of the EN.520.445 or EN.520.645 is beneficial.

Spring 2013

(4.38)

Spring 2014

(4.31)

Spring 2013

Professor: Hynek Hermansky

(4.38)

The best aspect of the course was the practical and hands-on learning experience. Students were given flexibility to work at their own pace and to do projects of their choosing. The worst aspect of the course was the independent assignment. The course would improve if the students had more support and resources from the professor, in addition to a clearer and better structured course syllabus. Prospective students should know that the course is independently driven and they wil get what they put into the course.

Spring 2014

Professor: Hynek Hermansky

(4.31)

The best aspect of this course seemed to be learning the methods of speech recognition. They also had lots of hands-on experience. But all topics are not covered in depth and it was not well organized, according to some students. It was suggested that more emphasis be put on state-of- the-art work in speech processing and that there be more assignments. Prospective students should have a background in speech processing and be interested in it.