ABSTRACT
In this chapter, the authors critically examine selected areas of research relevant to second language (L2) pronunciation instruction. The authors consider the issue of pedagogical norms to be separate from the issue of imagined interlocutors. In much computer-aided pronunciation teaching (CAPT) research, pronunciation is modeled by manipulating the frequency and/or salience of targeted pronunciation features, with learners often receiving feedback on their productions; the authors illustrate three strands of this work, all using freely available technology. In the first strand, visual information about pronunciation is often included in instructional materials to raise learners' awareness about targeted features or to provide unambiguous feedback about their pronunciation. The second strand incorporates the use of high-variability phonetic training (HVPT), which has a long history in speech research; it is based on theoretical views of how humans form categories and learn from examples. In the third strand, researchers have used internet-enabled technologies for learners to practice and give each other feedback on pronunciation.
