Quantitative Correlates as Predictors of Judged Fluency in Consecutive Interpreting: Implications for Automatic Assessment and Pedagogy

Abstract

This chapter presents an experimental study of consecutive interpreting which investigates whether: (a) judged fluency can be predicted from computer-based quantitative prosodic measures including temporal and melodic measures. Ten raters judged six criteria of accuracy and fluency in two consecutive interpretations of the same recorded source speech, from Chinese ‘A’ into English ‘B’, by twelve trainee interpreters (seven undergraduates, five postgraduates). The recorded interpretations were examined with the speech analysis tool Praat. From a computerized count of the pauses thus detected, together with disfluencies identified by raters, twelve temporal measures of fluency were calculated. In addition, two melodic measures, i.e., pitch level and pitch range, were automatically generated. These two measures are often considered to be associated with speaking confidence and competence. Statistical analysis shows: (a) strong correlations between judged fluency and temporal variables of fluency; (b) no correlation between pitch range and judged fluency, but a moderate (negative) correlation between pitch level and judged fluency; and (c) the usefulness of effective speech rate (number of syllables, excluding disfluencies, divided by the total duration of speech production and pauses) as a predictor of judged fluency. Other important determinants of judged fluency were the number of filled pauses, articulation rate, and mean length of pause. The potential for developing automatic fluency assessment in consecutive interpreting is discussed, as are implications for informing the design of rubrics of fluency assessment and facilitating formative assessment in interpreting education.