Developing a system for searching a shop name on a mobile device using voice recognition and GPS information
SeonYeong Kim, Sunghwan Kim, Hwan-Gue Cho · 2012
Although speech recognition is a most intuitive interface, the domain of speech recognition software has been limited until now. However, with the explosive growth of mobile devices such as smart phones and car navigation systems, the applicable range of voice recognition is widening. Furthermore, most speech recognition software manufacturers provide their mobile applications for free. Due to these kinds of support, it is more efficient for users to access specific information through a speech search; however, this method is generally avoided by users because of its inaccuracy. Voice recognition software packages provided by major companies such as Nuance and Google show high-accuracy speech searches in silent spaces (>90%). However, when users run a speech-based search in a noisier location such as a crowded street or inside a car, the accuracy is about 180 shop names in our experiments under 5 classes based on different noise levels. The accuracy of the experimental results was >90%, demonstrating that the results were almost fully corrected.