Abstract
High Variability Pronunciation Training (HVPT) is a highly
successful alternative to ASR-based pronunciation training. It
has been demonstrated that HVPT is effective in teaching the
perception of non-native phonemic contrasts, and that this skill
generalizes to the perception of unfamiliar words and talkers,
transfers to pronunciation, and is retained long-term. HVPT is,
however, not efficient and hence not motivating for the
learner. In this study, we therefore compare HVPT with an
alternative, namely oddity discrimination training. This
comparison, in which Mandarin-Chinese speakers were trained
to pronounce the English /r/-/l/ phonemic contrast, provides
preliminary evidence to support the use of discrimination tasks
in addition to identification tasks to add variety to HVPT.
successful alternative to ASR-based pronunciation training. It
has been demonstrated that HVPT is effective in teaching the
perception of non-native phonemic contrasts, and that this skill
generalizes to the perception of unfamiliar words and talkers,
transfers to pronunciation, and is retained long-term. HVPT is,
however, not efficient and hence not motivating for the
learner. In this study, we therefore compare HVPT with an
alternative, namely oddity discrimination training. This
comparison, in which Mandarin-Chinese speakers were trained
to pronounce the English /r/-/l/ phonemic contrast, provides
preliminary evidence to support the use of discrimination tasks
in addition to identification tasks to add variety to HVPT.
Original language | English |
---|---|
Title of host publication | SLaTE 2009 |
Publication status | Published - 2009 |