Current automatic handwriting text image recognition systems are far from being perfect and, in general, human intervention is required to check and correct the results of such systems. This is both inefficient and uncomfortable to the user. As an alternative to this post-editing process, a multimodal interactive approach is proposed, where user feedback is provided by means of touch-screen pen strokes and/or more traditional keyboard and mouse operation. Experiments suggest that, using this approach, significant amounts of user effort can be saved with respect to the conventional, non-interactive, post-editing process.