Digital Signal Processing Reference
In-Depth Information
different kinds of interface systems. The first system is a human navigator‚
who sits in a special chamber inside the vehicle and the navigator converses
and naturally. Second one is a wizard of Oz (WOZ) type system. The final
one is an automatic dialog set-up based on automatic speech recognition
(ASR). As it is normally done in many Japanese projects‚ we have employed
Julius [3] as the ASR engine. In Table 1-2 we tabulate the driver age
distribution.
Each subject has read 50 phonetically balanced sentences in the car while
the vehicle was idling and subsequently drivers have spoken 25 sentences
while driving the car. While idling‚ subjects have used a printed text posted
on the dashboard to read a set of phonetically balanced sentences. While
driving‚ we have employed a slightly different procedure for safety reasons.
In this case‚ subjects are prompted for each phonetically sentence from a
head-set utilizing a specially developed waveform playback software.
The recording system in our data collection vehicle is custom-designed
equipment developed at CIAIR for this task. It is capable of synchronous
recording of 12-channel audio inputs‚ 3-channel video data‚ and various
vehicle related data. The recording system consists of eight network-
connected computers‚ a number of distributed microphones and microphone
amplifiers‚ a video monitor‚ three video cameras‚ a few pressure sensors‚ a
differential-GPS unit‚ and an uninterruptible power supply (UPS).
