 |
INTERSPEECH 2013
|
 |
|
INTERSPEECH 2013
|
|
Two series of Electroencephalogram (EEG) measurements indicated that average auditory event-related potentials (ERPs) elicited by the disyllabic sound /asu/ were different between Japanese and English native listeners. Significant differences were observed in the time window where the P1-N1-P2 complex for /a/ is expected. This difference may be due to the absence/presence of the Auditory Change Complex (ACC) elicited by /s/. Furthermore, by combining the P1-N1-P2 complex elicited by each component of /asu/ (/a//s//u/) recorded individually, it was possible to compose ERPs similar to those elicited by /asu/. Interestingly, the optimized weights of /s/ were significantly lower for Japanese than for native- English listeners, suggesting that the trace of the ACC associated with /s/ was less visible in the actual ERP response to /asu/ for Japanese native listeners. These results may together suggest that Japanese and English native individuals process the /s/ in /asu/ differently and that the ACC is sensitive to language-specific perceptual categories.
Bibliographic reference. Sadakata, Makiko / Spyrou, Loukianos / Shingai, Mizuki / Sekiyama, Kaoru (2013): "Composing auditory ERPs: cross-linguistic comparison of auditory change complex for Japanese fricative consonants", In INTERSPEECH-2013, 901-905.