Fonetiek en vreemde-talenonderwijs
  • ISSN 0169-7420
  • E-ISSN: 2213-4883
Buy:$35.00 + Taxes


Speech rate variations often influence scores obtained in tests that aim at establishing a second language learner's listening proficiency. In this study a method will be proposed that may be immune to such variations. This method involves the construction of short speech samples varying in length from one to about ten words. There is evidence from the literature on the intelligibility of such speech fragments that samples of comparable durations are equally intelligible despite differences in speech rate: the sloppy articulation in fast utterances is supposed to be compensated by covering relatively more context.An account is given of two experiments set up to investigate the value of this claim. Results demonstrate that effects of speech rate variations are negligible for samples with average syllable durations beyond 132 milliseconds.Further reports are presented on two investigations into de discriminatory power of the testing procedure proposed. Fragments of speech read out in the second language turn out to be less intelligible than samples read out in the mother tongue. The amount of difference between scores under the two language conditions appears to be a function of the listeners' familiarity with the second language. It is concluded that the method put forward meets the essential requirement of discriminating in listening proficiency between mother tongue and second language and between levels of proficiency as well.Two final experiments were carried out to get more insight into the question of how the fragments might be presented. Voice adaptation effects are shown to play a role in the recognition of speech fragments. It can be inferred that listeners who are first presented with a passage read out by the speaker who pronounces the samples score higher than those who do not get the opportunity to get accustomed to his voice. Consequently it seems sensible to familiarize listeners extensively with the speaker's voice beforehand in order to avoid negative effects of voice adaptation.


Article metrics loading...

Loading full text...

Full text loading...

  • Article Type: Research Article
This is a required field
Please enter a valid email address
Approval was successful
Invalid data
An Error Occurred
Approval was partially successful, following selected items could not be processed due to error