Computer Assisted Speech Perception Evaluation & Training

According to San Diego State University's ROHAN Academic Computing, computer-assisted speech perception testing and training at the sentence-level (CASPERSENT) is "designed for the evaluation of auditory, visual and auditory-visual speech perception performance at the sentence level." It is an experimental approach, funded by the US Department of Education, to test sentence comprehension among adults with acquired hearing loss and to find out if sentence perception can be improved with computer-assisted training.

Self-Scoring

• In the self-scoring assessment, a subject is shown a person speaking a sentence. The subject repeats a sentence and then clicks a button to reveal the text of the sentence. The subject then clicks on the button to indicates the words that were correctly repeated.

Other-Assisted

• In other-assisted assessments, a subject repeats a sentence and the helper (frequently a clinician) then reviews and scores the text. The text is then shown to the subject with the missed words omitted and given another try at the sentence.

Purpose

• According to ROHAN, the purpose of these tests is to provide a way of assessing word recognition in normal conversation. It can show, among other things, holes or flaws in comprehension abilities. The hope is that, over time, this training will improve comprehension.