Perceptual evaluation is a conventional method for studying typical and atypical production of speech. It is also the primary tool that speech and language therapists use to assess speech disorders, monitor treatment progress, and evaluate treatment outcome. Although auditory-perceptual based procedure is the main method used in speech evaluation, visual examination of the movement of articulators is also important, as it can provide essential information that adds to the understanding of the nature of the speech difficulties as well as any compensatory strategies that a speaker might use to reduce the impact of speech disorders on speech clarity. Auditory-perceptual evaluation usually involves listeners listening to a person’s speech and judging the severity level of features present (e.g., hoarseness, hypernasality, imprecise consonants) or transcribing the correct and incorrect speech sounds produced. We, including those who have not received relevant professional training, are able to carry out perceptual evaluation of speech. However, the performance may vary in terms of how accurate the judgment or phonetic transcription is, and how consistent it is within and between the listeners. There are various factors that can influence the performance, hence, a way to enhance it is to control those factors for as much as possible and to train listeners in carrying out perceptual evaluation of speech. In this talk, we will look at a framework that comprises the components involved in perceptual evaluation of speech, with focus on listening training that draws on knowledge from areas such as perceptual learning, speech corpora. Gaps of knowledge and ideas for further research in the broader topic of effective listener training for perceptual evaluation of speech will be discussed.