Digital Signal Processing Reference
In-Depth Information
therefore this should be included in the speaker modeling phase. Previous
studies have clearly shown that the effects of speaker stress and Lombard
effect (i.e., speaking in noise) can cause speech recognition systems to fail
rapidly[16]. In addition, microphone type and placement for in-vehicle speech
collection can impact the level of acoustic background noise and ultimately
speech recognition performance. Figure 2-1 shows a flow diagram of the
proposed CU-Move system. The system consists of front-end speech
collection/processing tasks that feed into the speech recognizer. The speech
recognizer is an integral part of the dialogue system (tasks for Understanding,
Discourse, Dialogue Management, Text Generation, and TTS). An image of
the microphone used in the array construction is also shown (Figure 2-2). The
back-end processing consists of the information server, route database, route
planner, and interface with the navigation database and navigation guidance
systems. Here, we focus on our efforts in multi-channel noise suppression,
automatic environmental characterization, robust speech recognition, and a
proto-type navigation dialogue.
Figure 2-1.
Flow Diagram of CU-Move Interactive Dialogue System for In-Vehicle Route
Navigation
