6.2 Gesture Tier(s)

Each gesture tier contains a sequence of intervals of all gestures produced with one particular body part (or a pair of body parts) and the corresponding CoGesT transcription symbols. That is, gestures with hand, head, torso etc. are all annotated on separate tiers.

**Figure 5:** *Example of a two-tiered annotation with the TASX-Annotator.*
$\includegraphics[scale=0.5]{tasx_video.eps}$ $\includegraphics[scale=0.36]{tasx_menue.eps}$

Figure 5 illustrates a two-tier annotation in the telling of a story. The top tier gives an orthographic transcription of the words, and in the bottom tier the gestures of the hand are transcribed. The TASX-environment (Milde and Gut, 2001,2002) offers an XML-based corpus environment for the annotation and analysis of multimodal data.

Thorsten Trippel 2003-08-12