Digital Signal Processing Reference
In-Depth Information
Figure 1-8.
Partial dialogue sequence map in the Layered Intention Tag (LIT) Structure.
5.2
System Performance Differences between Human and
WOZ Navigators
As discussed earlier‚ we have collected in-car information conversations
using an actual human navigator‚ Wizard of Oz (WOZ) setup‚ and speech
recognition (ASR) system. Since the ASR systems are configured to work in
the system initiated conversation mode and they require considerable training
to learn the speaking habits of drivers -more than 800 in our case- they were
thought to be highly restricted this application. On the other hand‚ the human
navigator and the WOZ setups have been observed to be very effective and
easy to realize. Hence‚ we will be presenting some comparative results on
driver behaviour if they converse with a human navigator or the Wizard of
OZ system.
In Figure 1-9‚ we have plotted the performance of top ten layered intention
tags (LIT) in the restaurant query task‚ where lines represent the number of
phrases per speech unit in the case of human navigator (H) and the WOZ
system (W). We have also included a bar diagram for the occurrence rate of
linguistic fillers.
