The effect of auditory and articulatory phonetic training on the perception and production of l2 vowels by catalan/spanish learners of english
- Autores: Cristina Aliaga García
- Directores de la Tesis: Joan Carles Mora Bonilla (dir. tes.)
- Lectura: En la Universitat de Barcelona ( España ) en 2017
- Idioma: español
- Tribunal Calificador de la Tesis: Julio Cebrián (presid.), Anabela Rato (secret.), Jane Setter (voc.)
- Programa de doctorado: Programa de Doctorado en Estudios Lingüísticos, Literarios y Culturales por la Universidad de Barcelona
- Materias:
- Enlaces
- Tesis en acceso abierto en: TDX
- Resumen
Adult second language learners often experience major difficulties in perceiving and producing non-native speech sounds. Several perception training studies (Iverson & Evans, 2007; Nishi, & Kewley-Port, 2007; Carlet & Cebrian, 2014) have shown that secondlanguage (L2) learners can improve their L2 perception, also demonstrating significant gains in L2 production (Bradlow, Pisoni, Akahane-Yamada, & Tohkura, 1997; Kartushina et al., 2016). However, research on the assessment of methods other than perceptual training for non-native vowels is still scarce, and none of the previous vowel studies has compared the impact of auditory vs. production-based training on a full set of vowels. The purpose of this study was to evaluate two training methods that might be used to improve learners’ identification and articulation of the 11 English RP monophthongal vowels (/i: ɪ e ɜ: æ ʌ ɑ: P N9 ʊ u:/).
A total of 84 bilingual Catalan/Spanish learners of English were divided into two experimental groups and a control group, and all were tested on vowel identification, identification of synthesized vowels (with manipulated duration), vowel discrimination and vowel production based on a delayed repetition task. Two groups of bilingual Catalan/Spanish learners of English (N=64) were assigned to different types of audiovisual High Variability Phonetic Training (HVPT) based on natural CVC words from multiple talkers, either identification (ID) or articulatory (ART) training. Both training procedures comprised 10 one-hour computer-based sessions over 5 weeks, which guaranteed exposure to a minimum of 132 trials/ session. Whereas the ID training required learners to focus on the critical audiovisual cues to recognize the vowel category within a vowel subset, ART training learners were expected to focus on the relevant audiovisual cues for more accurate vi vowel articulation. Auditory feedback provided assistance to correct identification, or to change erroneous articulations. This paper compares some remarkable effects of perceptual and production-based audiovisual HVPT on the perception and production of the fullset of English vowels.
The two HVPT groups showed higher accuracy in vowel perception, as well generalization to new words, talkers and contexts. HVPT not only improved vowel identification and discrimination, but also reduced the learners’ heavy reliance on vowel duration and improved their use of spectral cues in English vowel perception. However, a clear advantage of the ID group was seen in a better identification of trained words and a lesser degree of error dispersion per vowel. Both HVPT methods were effective in leading to significant formant movement for some vowels, with less spectral overlap, but differences in the amount of spectral shift after each training method suggest that ART training was more effective in vowel production. Training was effective in making the production of contrastive vowels more distinct and revealed a conscious attempt of learners to produce acoustically distinct vowel quality targets, with a great deal less spectral overlapping.
Pedagogical implications will be discussed.
