The present study examined the effects of using automatic speech recognition (ASR) technology on oral complexity in a flipped English as a Foreign Language (EFL) course. A total of 160 undergraduates were enrolled in a 14-week quasi-experiment. The experimental group (EG) and the control group (CG) were taught with a flipped approach, but the EG students needed to undertake an additional pre-class task with ASR technology. In each unit, all students’ in-class task performance was recorded, based on which the metrics of oral complexity were coded and computed. A two-way between- and within-subjects repeated measures design was conducted to examine the effects of the group factor, the time factor and the group × time interaction effects. The results showed that the EG students performed statistically better than their counterparts in the CG on lexical complexity and syntactic complexity. Moreover, significant improvement in phrasal complexity was witnessed over time in both groups. Significant group × time interaction effects were witnessed on overall complexity or subordination complexity. The gradients of the EG trajectories of the two metrics were greater than those of the CG. However, on phrasal complexity, the interaction effect was not significant.

Although the automatic speech recognition (ASR) technology is increasingly used for commercial purposes, its impact on language learning has not been extensively studied. Underpinned by the sociocultural theory, the present work examined the effects of leveraging ASR technology to support English vocabulary learning in a tertiary flipped setting. A control group and an experimental group of college students participated in a 14-week study. Both groups had their English classes in a flipped fashion, but the experimental group was assigned with ASR-assisted oral tasks for pre-class self-learning. The pre- and post-intervention in-class task performance of both groups was audio-recorded and transcribed for data analysis. The triadic complexity-accuracy-fluency (CAF) framework was adopted to evaluate the participants' vocabulary learning. The between- and within-subjects effects were examined mainly through procedures of MANCOVA and mixed-design repeated measures ANCOVA. Results showed that on all the metrics of lexical complexity and speed fluency, the experimental group outperformed the control group, and had significant growth over time. On the other hand, the control group only improved significantly overtime on the G-index. On lexical accuracy, there was no significant difference between the two groups, and the within-subjects effect was not significant for either group. The findings lent some support to Skehan's Trade-off Hypothesis and discussions were conducted regarding the triarchic CAF framework.

Background While automatic speech recognition (ASR) is increasingly used for commercial purposes, its influence on the learners' linguistic performance in terms of oral complexity, accuracy and fluency was under‐explored. To date, few studies have been conducted to investigate how the dictation ASR technology could be incorporated into language classrooms to facilitate the learners' language acquisition. Objectives This study aimed to examine the effects of ASR‐based technology on English learners' oral accuracy and fluency and depict the corresponding development trajectories. Methods A total of 160 first‐year university students were enrolled in a 14‐week quasi‐experiment based on a longitudinal research design. Both treatment and control groups were taught with the flipped classroom approach, but the students in the treatment group were particularly required to undertake a pre‐class task with ASR technology. Students' Unit Task performance was audio‐recorded, and the metrics of oral accuracy and fluency were coded and computed based on the recording transcripts. A two‐way repeated measures ANCOVA was conducted to investigate the between‐ and within‐subjects effects as well as the corresponding interaction effects. Results In terms of the between‐subjects effect, the treatment group outperformed the control group on phonological accuracy, speed fluency and repair fluency. In terms of the within‐subjects effect, significant gains in lexical and morphosyntactic accuracy were witnessed over time in both groups, but the performance of the treatment group tended to be more stable. On all the fluency metrics and phonological accuracy, no significant within‐subjects improvement was seen over time. Implications The development of oral accuracy may generate a negative impact on that of fluency. Therefore, course developers and teachers need to design special tasks and provide conditions conducive to the development of the students' oral fluency. Moreover, the mobile‐based dictation ASR application could incorporate adaptive artificial intelligence, which may escalate the students' fossilization of learner interlanguage.