Purdue signature

May 8, 2014

Study: Fluency outweighs pronunciation for understanding non-native English speakers

WEST LAFAYETTE, Ind. — Pronunciation accuracy may not be the most important thing for making non-native English speakers easier to understand, but rather it is their fluency, including fewer pauses, restarts and speech rate, according to research from Purdue University.

"We found that people who speak English as a second language were more likely to be judged as easy to understand when they spoke with fluency, regardless of the accuracy of their pronunciation," said Alexander L. Francis, an associate professor of speech, language and hearing sciences who studies speech perception and cognitive hearing science. "With more fluent speech, listeners are not working so hard to keep track of what the speaker is trying to say, so they can devote more effort to figuring out sounds the speaker is trying to produce. These findings could mean a new approach for second-language instruction and assessment, but more study is needed and we are taking a closer look at the difference fluency makes."

Francis and doctoral student Mengxi Lin presented their findings Tuesday (May 6) at the annual meeting of the Acoustical Society of America in Providence, Rhode Island. The paper related to the presentation is available online and it will be submitted for journal publication.

The findings are based on the speech of 20 non-native speakers of English whose first language was Chinese or Korean. Their audio speech samples, which varied from low to high proficiency, were played for native English speakers. In the first group of English-speaking listeners, they ranked the speech samples based on acceptability, intelligibility and understanding. In the second group, the listeners participated in a listening task to objectively assess the speakers' capability of being understood. The data also was assessed for speech rate, pause duration and sound properties related to pronunciation accuracy.

"Our results suggest that achieving proficiency may depend more on developing fluent speech patterns and less on attaining pronunciations that are perceived as native-like to native speakers," said Lin who is studying linguistics. "These findings may sound counterintuitive, but as the speakers with better fluency in this study also received higher subjective ratings of intelligibility and acceptability and lower ratings of listening effort, it suggests that native listeners are better able to cope with divergent pronunciations if they appear within otherwise fluent speech."

The researchers are further studying this by evaluating the role played by fluency's characteristics such as speech rate and frequency of pauses. While they are evaluating the speech of native speakers of East Asian languages, they believe the results would be the same for other languages as well.

"If it is the case that fluency is the most significant factor, then poorly fluent speakers from any language should fare worse than highly fluent speakers from the same language," Francis said. "But this would need to be tested to know for sure."

This was study was funded by Purdue's linguistics program and the Purdue Research Foundation. 

Writer: Amy Patterson Neubert, 765-494-9723, apatterson@purdue.edu

Sources: Alexander Francis, francisa@purdue.edu

Mengxi Lin, lin211@purdue.edu 

Related websites:

College of Health and Human Sciences

College of Liberal Arts

Acoustical Society of America press room 


The relationship between fluency, intelligibility, and acceptability of non-native spoken English

Mengxi Lin and Alexander L. Francis

Non-native accented speech is typically less intelligible and less fluent than native speech, but it is unclear how these factors interact to influence perceived speech quality. To investigate this question, the speech of 20 non-native speakers of English varying in proficiency and native language was evaluated. Subjective measures of speech quality (listening effort, acceptability and intelligibility) were compared to objective measures of word recognition by native listeners, and to acoustic measures of fluency and of segmental and suprasegmental properties related to intelligibility. Results showed that subjective quality measures were highly related to one another and to word recognition, and were most strongly predicted by measures of fluency. Segmental and suprasegmental measures did not predict word recognition or subjective speech quality. There was also an interaction between the effects of proficiency and speaker's native language on word recognition, but this did not extend to subjective measures. Finally, listeners who first heard high-proficiency speakers gave overall lower subjective quality ratings but there was no interaction between proficiency and presentation order. Multivariate analyses suggest that factors related to speaking rate, including pause duration, have the greatest effect on measures of acceptability, intelligibility and listening effort.