Effects of training, style, and rate of speaking on speech perception of young people in reverberation
Nao Hodoshima, Takayuki Arai, Kiyohiro Kurisu · The Journal of the Acoustical Society of America · 2008
Because of the difficulty of listening to speech in reverberation (e.g. train stations or airports), we need to find characteristics of intelligible speech sounds that are appropriate for announcements over loudspeakers in public spaces. This study investigated the effects of training (seven talkers who have received speech training or not), style (conversational/clear) and rate (normal/slow) of speaking on speech perception of young people in simulated reverberant environments. The talkers were instructed to speak nonsense words embedded within a carrier sentence clearly or normally in an anechoic room, and listening tests were carried out with young people in simulated reverberant environments. Results showed that correct rates significantly differed among the talkers, but no difference in correct rates was found between the two speaking rates, and conversational speech had significantly higher correct rates than clear speech. Casual inspections of the stimuli indicate that vowels are enhanced as well as consonants in clear speech so that clear speech had lower correct rates than conversational speech due to increased reverberant masking in clear speech compared to conversational speech. [Work supported by Sophia University Open Research Center from MEXT.]