The Analysis on Commercial and Open Source Software Speech Recognition Technology

Abstract

The speech recognition technology is one that converts acoustic signals obtained through a sound sensor such as a microphone into words or sentences. The speech recognition technology appeared in the 1950s and has been studied continuously, but it has not been popularized until the mid-2000s because of its low speech recognition rate. Recently, however, the speech recognition technology, which has been used limitedly only for specific applications so far, is being developed rapidly along with the proliferation of portable computing terminals represented by smart phones and the expansion of cloud infrastructure to support them. In particular, the speech interaction system combining this technology with artificial intelligence is attracting attention as a next generation interface, and has been used in various fields such as smart phones, smart TV and automobiles. Recently, Samsung Electronics has released ‘Bixby’ combining artificial intelligence and speech recognition, and a lot of companies such as Google and Naver have offered speech recognition technologies as open API. This paper compared and analyzed features and functions of typical speech recognition software in the commercial and open source software fields. It is expected that the comparison and analysis on the features and functions of commercial and open source software in the speech recognition software field carried out in this study could suggest a standard for selecting speech recognition technologies to help use proper API in context.