<p>This paper presents a large-scale study of the duration of the first two segments in words beginning with the orthographic sequence &lt;irr-&gt; (e.g., <i>irrational, irremovable</i>) using speech data from British English and American English. We investigate the effects of several predictors encoding the extent to which these words are segmentable into their constituent parts using a dataset of over 1,500 audio extracts. When using categorical predictors to code segmentability, we find effects that are consistent with the literature: more segmentable words have longer segments in the prefix and at the prefix-base boundary. We also report mixed results regarding measures of frequency as proxies of segmentability, in line with the available literature. We also report effects of non-morphological factors: [ɹ] is overall longer in American English than in British English, and we find longer [ɹ] durations for male speakers than for female speakers.</p>

错误:搜索内容不能为空,请输入英文关键词
错误:关键词超出字数限制,请精简
高级检索

Morpho-phonetic effects in <irr-> prefixed words in British and American English

  • Quentin Dabouis,
  • Olivier Glain,
  • Sylvain Navarro

摘要

This paper presents a large-scale study of the duration of the first two segments in words beginning with the orthographic sequence <irr-> (e.g., irrational, irremovable) using speech data from British English and American English. We investigate the effects of several predictors encoding the extent to which these words are segmentable into their constituent parts using a dataset of over 1,500 audio extracts. When using categorical predictors to code segmentability, we find effects that are consistent with the literature: more segmentable words have longer segments in the prefix and at the prefix-base boundary. We also report mixed results regarding measures of frequency as proxies of segmentability, in line with the available literature. We also report effects of non-morphological factors: [ɹ] is overall longer in American English than in British English, and we find longer [ɹ] durations for male speakers than for female speakers.