In this research, we integrate the voice commands technique into Google Map. It means
that we can control part of the movements for Google Map search without using the mouse or
keyboard but with voice. Our voice command system is built on the client side. The biggest
different between our system and state-of-the-art real-time speech processing system is that
all the computation about the speech process always work on the client side. For our corpus,
we choose the Top100 scenic spots in Taiwan and some specific control commands as our
training data. In the experiment of our research, we make use of the different ways to train
the acoustic models and design dictionary and language models to estimate the efficiency on our system. Actual usage in the system, we can move the map center to the specific location sequentially by voice command operations for location, control and coordinate. we estimate the overall search process time on some specific locations by different users. It spends 20.8 seconds in average which spends most of time in recording stage.
| Identifer | oai:union.ndltd.org:NSYSU/oai:NSYSU:etd-0518112-140702 |
| Date | 18 May 2012 |
| Creators | Wu, Po-feng |
| Contributors | Liang-Chih Yu, Chung-Hsien Wu, Chia-Ping Chen, Hsin-Min Wang |
| Publisher | NSYSU |
| Source Sets | NSYSU Electronic Thesis and Dissertation Archive |
| Language | Cholon |
| Detected Language | English |
| Type | text |
| Format | application/pdf |
| Source | http://etd.lib.nsysu.edu.tw/ETD-db/ETD-search/view_etd?URN=etd-0518112-140702 |
| Rights | user_define, Copyright information available at source archive |
Page generated in 0.0065 seconds