Voice Recognition Competition
Google, Apple, Microsoft, and Amazon are scrambling to develop voice recognition technology. The tech is receiving a lot of attention as a core method to control next-gen smart devices like smart watches.
Google revealed a Korean version of its new voice-based features such as voice answering and action functions on Dec. 4. Smartphones or smart watches with a voice answering function can also reply with their own voices, as opposed to the previous method that merely recognizes a voice and does a search. The voice action feature is a service that delivers what was said via text messages.
Amazon unveiled its voice recognition device called Echo in November. The soda can-sized device can understand and deliver the meaning of a word from the context thanks to its communications feature.
Apple is a dominant player in the voice recognition area. After showcasing its speech recognition software Siri in 2011, the company has continued to improve it. The intelligent personal assistant can give real-time information about weather or other information through voice commands. Simple communication is possible as well.
Local companies are also stepping up efforts to develop voice recognition technology. Naver Labs is involved in the development of a voice recognition method that can best understand the Korean language. Currently, its trial edition is available, and it continues to improve.
Last month, Daum Kakao added a text-to-speech feature to speech synthesis program Newtone Talk. It is possible to convert newly-coined words into voice. The firm made public the technology so that other developers can create new services.
An advance in voice recognition technology is deeply related to the Internet of Things era. Voice is considered to be the best method to control smartphones, PCs, and other devices such as TVs, refrigerators, and lighting. Unlike a keyboard or a mouse that needs a screen to operate, only a small microphone is necessary to run a voice recognition-based system.
The Korea Institute of Science and Technology Information predicts that the global voice recognition market will increase from US$72 billion (80 trillion won) this year to US$113 billion (126 trillion won) in 2017.