Speak Code is a project that aims to let people program and navigate through code using only their voice. Speak Code is a code editor that is connected to a local (needs no internet) ASR engine that is also a part of the project. The user interacts with the editor through pre-defined commands. These commands are in English words and are designed to be as close as possible to the common programming languages keywords, such as “while”, “for”, “if”, etc. These commands needs to be memorized by the user in order to achieve a coding speed close to that of regular typing.
Programming, as a profession or a hobby, has been adapted for every area of work and art as programming started to be effective since the 1940s. Nearly everything is now being enhanced with the help of computer programs in every field of work we can imagine, they are used in teaching at schools, medicine, agriculture, banking systems, space exploration and much more.
With the increasing variety in computer programs, the continuously increasing complexity of the programs and its criticality, programs are written in millions of lines of code and even billions in large companies like Google; the Google codebase includes approximately one billion files and has a history of approximately 35 million commits spanning Google's entire 18-year existence. Consequently, the number of programmers is increasing day by day and the fields of work are drastically expanding.
RSI is a form of overuse injury caused by repeating the same motions for hours on end over extended periods of time, especially in the wrists and fingers. It was reported that 1.8 million workers are afflicted by RSI per year. Also, half of the people who work with computers suffer from RSI symptoms.
Taking into consideration the physical health limitations of programmers, how they may handicap the skills of programmers and stand as an obstacle in front of the increasing demands of programs, we present here Speak Code; a software solution that allows users to program using their voice only.
The research in voice and speech recognition has started since the 1950s and is still emerging. A Stanford study revealed that speech recognition is now about three times as fast as typing on a cell phone and that the word error rate has dropped significantly.