Measurement of Usability for Speech Recognition on Ok Google

摘要

As technology develops, Google can accept input not only by typing keywords but also through speech. Speech as input to Google is known as the Ok Google feature. However, many search engine users never pay attention to search engine definitions. So, websites are only limited to projects without further management, and that is what happens to the current website. Therefore, in this study, an assessment of usability is conducted on Google especially in the Ok Google feature using Nielsen Model. The research method used in this research is the quantitative research approach, a method which includes approach and kind of research, population and sample, instrument, data collection technique, and data analyze. Data collection techniques include learnability, efficiency, memorability, errors and satisfaction assessment and Likert scale ranges from 1 to 4. It can be signed into categories of sites with usability values, namely "very difficult to understand," "difficult to understand", "easy to understand" and "very easy to understand". The construct validity is with Orthogonal Iterations with Index reliability of 0.91. The results of this study are the usability testing of Ok Google tends to be significance (α) of 5% with data of 124 is 1.66, feedback statement in the questionnaire is considered valid, then the questionnaire results show α = 0.91 which means the measurement of usability on Ok Google Feature is an excellent level. The average usability value obtained by Ok Google in Indonesia is included in the category of usability "easy to understand". In addition, it was found that Ok Google has the highest memorability level and the lowest error.