Age estimation in foreign-accented speech by first and second language speakers

Abstract

Previous research has shown that listeners are fairly accurate in estimating speakers’ age from their speech (e.g. Moyse, 2014). Some research has explored the effect of the speaker’s first language (L1) on age estimation. Nagao and Kewley-Port (2005) presented English and Japanese stimuli to two groups of listeners from these L1s and found that listeners were more accurate at estimating age in the familiar language. Rodrigues and Nagao (2010) extended this line of research to foreign-accented speech by playing English language clips recorded by native speakers of Arabic and English to native English listeners with more and less experience with foreign accents. They found a higher correlation between estimated and chronological age for English speakers than for Arabic speakers and a higher correlation for the more experienced listeners than for the less experienced ones for Arabic speakers but not for English speakers.

The current study sets out to compare age estimation accuracy by first and second language speakers. The audio stimuli were 40 clips of 20 British English speakers and 20 Japanese L1 speakers (age range: 18 - 71) reading a passage in English. Two groups of listeners (36 English and 23 Japanese L1 speakers) were presented with the audio-stimuli and were asked to estimate the speakers’ age.

Statistical analysis shows that male speakers were estimated to be older than females by both English and Japanese listeners and Japanese L1 speakers were estimated to be younger than their English counterparts by English L1 listeners only (Figure 1). This variation can be explained by listener familiarity with the language or accent: English listeners exhibited the highest correlation of 0.64 for English speakers, followed by Japanese listeners performing equally well with English- and Japanese-accented English (0.44 and 0.45), and English listeners listening to Japanese-accented English coming last (0.37). These findings have theoretical implications as they highlight that even such a seemingly universal phenomenon as age may be expressed and perceived differently by people from different language backgrounds and of varying familiarity with languages and accents. This further supports previous studies that show a connection between age and sociolinguistic features, reflecting that age is expressed both physiologically and socio-culturally. The practical implications include our need for awareness of such differences when age estimation occurs in real life (e.g. forensics).