With the popularity of mobile and personal assistants like Siri and Google Now, voice interfaces are increasingly commoditizing. Whether to make a phone call or to schedule a task without touching the screen, the passage through the voice is really something practical as more people are using voice recognition services.
Annayang is my absolute favorite speech recognition tool because it is Faster, more accurate, multi-language, just weight 3kb, no dependencies and free to use. It is a very famous voice recognition program.
Its use is really simple: just define patterns such as “hello” or “Run the command: cmd_name“, and to connect the corresponding functions. Whether to record a favorite resource, comment on an article or decide to put an article pannier, this tool really helps to execute orders within seconds, and can significantly improve the conversion of the site.
I also found two simple examples of using Annyang in Code pen.
Julius is working more like a speech recognition library for the web and build with Emscripten. You must be wondering what is the speech recognition library is so let me explain: browsers have some build in voice control tools, but they always depend on some sort of remote service like Google Now—which handles the basic voice processing from a distant server. This can slow things a little bit. So JuliusJs gets around this problem by running a speech recognition library inside the browsers.
Pomerantz develops this tool and it is an open source project so you guys also can contribute.
Compare to Annayang, Voicebox is a new project voice control app built with electron and NodeJS. It requires NodeJs to be installed in your system to work. For now this tool understands very limited amount of sentences, but it’s really easy to add new ones. I suggest you install it to take a look, it’s fascinating.