The Effectiveness of Pronunciation Training Software in ESL Oral Fluency Development

The study examined the effectiveness of automatic speech recognition – pronunciation feedback (ASR-PF) software in oral fluency training. The software employed for this purpose was NativeAccent v.3 - a pronunciation-training platform for the speakers of English as a second language. The experiment evaluated two groups of native speakers of Mandarin enrolled in various graduate programs at a large research university in the United States. All participants were enrolled in a course designed for international teaching assistants and emphasizing oral language production and pronunciation skills. Whereas the experimental group was exposed to NativeAccent v.3, the control group experienced no such exposure. The participants’ fluency was assessed using the measures of speech rate and silent pausing. Automated fluency evaluations were performed at the beginning and at the end of the study using Praat. In addition, participants’ performance was automatically evaluated by NativeAccent v.3. Due to its considerable limitations, the study showed modest influence of the software on the participants’ L2 oral fluency development. Still, despite these limited effects ASR-PF software appears to be a promising teaching tool with a lot of pedagogical potential.