English speech training using voice conversion

Effects of imitation and self-imitation practice on L2 pronunciation progress

“…Over the next few years there were a growing number of studies that focused on imitation and selfimitation experiments. Peabody and Seneff (2006) presented a pattern that previously appeared in Nagano and Ozawa's (1990) study. In order to teach L2 pronunciation more efficiently, they focused on tonal (Mandarin) and non-tonal (English) languages.…”

“…Again, similarly to Nagano and Ozawa (1990) and Peabody and Seneff (2006), Bissiri et al (2006) came to the same conclusion by investigating the process of L2 German pronunciation training on Italian speakers. Participants' own voices with corrected prosody resulted in the improvement of L2 pronunciation.…”

“…The significance of the current study lies in two facts. Firstly, it attempts to replicate to a certain degree research to corroborate or refute earlier findings, including Ding et al's (2019), De Meo et al's (2013, Felps et al's (2009) and Nagano and Ozawa's (1990). Secondly, it focuses on L2 English (stress-timed) and L1 Polish (a syllable-timed language) that, to the author's knowledge, has not yet been examined from the perspective of imitation and self-imitation tasks and their impact on L2 pronunciation skills.…”

“…On a larger scale, in their pioneering work Nagano and Ozawa (1990) used the prosodic-conversion method to evaluate the effectiveness of pronunciation training of Japanese learners. The method was based on the production of synthetic speech by converting such prosodic parameters as F0 contours and segmental duration of sonorants in subjects' speech into native English speakers' speech.…”

“…Starting from the 1990s, for the last few decades, the idea of automatic identification of foreign accents and the growing technological possibility to investigate the effects of students' self-imitation method using their own voices has attracted a large number of researchers who believe that prosodic (relating to the rhythm and intonation of language) and accent conversion can bring implicit, corrective and encouraging feedback to L2 pronunciation learners (e.g. Repp and Williams 1987;Nagano and Ozawa 1990;Markham 1997;Munro and Derwing 1999;Eskenazi 1999;Bissiri and Pfitzinger 2009;De Meo et al 2013;Pellegrino and Vigliano 2015;Ding et al 2019 etc. ).…”