Authors : S. Lokesh and G. Balakrishnan
Abstract: One of the draw back in Spoken Dialogue System through speech is the brittleness of Automatic Speech Recognition (ASR). ASR Systems often unpredicted the user input and they are unreliable when it comes to judging, recognition failures and lacking in estimating the own performance of an interaction system. Humans overtake ASR Systems on most tasks related to speech understanding. One of the reasons is that humans make use of much more knowledge. For example humans appear to take a variety of knowledge-based aspects of the current dialogue into account when processing speech. The main purpose of this study is to investigate whether speech recognition also can benefit from the use of higher level knowledge sources and dialogue context when used in human computer interaction. This review provides more insight into what type of knowledge sources in spoken dialogue systems would be potential contributors to the task of ASR and how such knowledge can be represented computationally. The purpose of this survey was also to exemplify the difficulties that arise when using speech in dialogue systems. Many of these difficulties have also been encountered in the experiments.
S. Lokesh and G. Balakrishnan, 2013. Analysis of Various Efforts to Compensate for Automatic Speech Recognition Deficiencies in Spoken Dialogue System. Asian Journal of Information Technology, 12: 7-13.