Voice coding, or voice programming

For some time now, we have been using voice commands that become commands for a computer or phone. Speaking commands can become programming commands. Is it possible to program hands-free and what is voice coding?

Alexa, Siri, Apple CarPlay and Android Auto – we use these voice commands to operate many gadgets. There are also mobile apps that respond to voice commands. There is also an option to “write” programming code by voice. This would seem impossible – after all, the primary tool of a programmer is a programming language, typed by hand.

Voice coding: what is it?

Voice coding is quite a challenge – it contains special characters and phrases that you need to know in order to speak them. There are many types of syntax and conventions and words outside the dictionary. In addition, the source code is a text that is more difficult to transcribe “by ear” than literary text – there are many parentheses, indentation at different levels, a large number of punctuation marks. Another thing is to take care of accuracy – a typo in the code will block further processing of the input data. As a result, the operation will fail.

Voice programming application

Voice programming is possible with the Serenade application, which relies on a speech-to-text engine that turns speech into programming code. In simple terms: when a programmer speaks a command aloud, the software engine passes it to its natural language processing layer. Subsequently, machine learning models identify and translate typical constructs into syntactically correct code, and in this way voice code is created. So it doesn’t work like Google Asistant, which processes speech, or more accurately, “translates” natural language into voice programming.

Voice coding: interesting facts

Serenade raised $2.1 million in funding in 2020 for further development. The app’s co-founder was reportedly so tired of writing code that he could no longer write. At one point, he faced an ultimatum: choose another career path or… a completely different solution. He chose the latter and created software that allows him to pursue his favorite field.

Hands-free programming

A similar story involved Talon platform developer Ryan Hileman. He left his full-time job as a programmer due to increasingly severe hand pain. Talon works on the principles of speech recognition, eye tracking and noise capture. The functioning of the platform is based on Facebook’s Wav2letter, and Hileman has expanded to include detection of programming commands. The application additionally features innovative mouse navigation (moving the cursor across the screen by moving your eyeballs and clicking by moving your mouth).

Voice coding: the future

Despite these innovations, it is hard to believe that hands-free programming will become the leading development in IT. Most programmers will stay with solutions that are more convenient for them. However, this niche has the potential to be exploited for those who, for some reason, cannot use traditional programming tools or …eager for new experiences.