In the traditional cascade modeling approach, automatic speech recognition (ASR) first produces a single text string, which is then passed to retrieval. Small transcription errors can change query ...
Abstract: Nowadays, geo-textual data, which consists of both spatial and textual information, have become increasingly preva-lent. The location-based services are ubiquitous in daily life to help ...
Some results have been hidden because they may be inaccessible to you
Show inaccessible results
Feedback