Aims of this study are: 1) identify respiratory features that could serve as objective markers of fluency of aloud reading; 2) investigate the effects of computer-assisted training on the improvement of speech-breathing coordination. Our training method combines the principles of repeated and assisted close-shadowed reading. Reading assistance takes over the principles of karaoke: highlighting text units in sync with pre-recorded adult performance. The aloud reading of 66 French young pupils is studied. They were divided into 3 groups: control, trained with word highlighting vs. trained with word and breath group highlighting. All children were recorded before and after 3 weeks of training while reading both a trained and an untrained text. The results indicated that respiratory planning and pauses management improved by computer-assisted reading training. However, there was no significant transfer of these improvements to the untrained text.