Over a million developers have joined DZone.
{{announcement.body}}
{{announcement.title}}

Do You Want to Speak to My Computer?

DZone's Guide to

Do You Want to Speak to My Computer?

The idea is that your phrases raise an event that executes some piece of code, facilitating you to create code by speaking in natural language to your computer.

· AI Zone ·
Free Resource

Did you know that 50- 80% of your enterprise business processes can be automated with AssistEdge?  Identify processes, deploy bots and scale effortlessly with AssistEdge.

CHI is an acronym that I have coined myself, and it translates into computer to human interface. The idea is that we need a Rosetta Stone that can help us translate from natural speech to computer code. Such an interface needs to be extremely customizable since each user has different needs. This implies that we'll need a dead simple programming language that any user can, at least to some extent, be able to write code in. This language needs to also be extremely agile, dynamic, and flexible, and much more closely resemble our existing communication protocols (language) than those we have available to ourselves today.

I believe Hyperlambda is that language.

In the video below, I demonstrate how to use natural English speech combined with speech recognition, speech synthesis, and Hyperlambda to initiate a semantic and meaningful discussion with my web server in order to have my web server evaluate some piece of Hyperlambda as a response to my phrases. If you wish to reproduce what I do in the video, you can visit my home server and click the magic wand icon in the top right corner. Note that speech recognition only currently works with Google Chrome.

To illustrate how this language much more closely resembles natural language, I have included a snippet of Hyperlambda below that creates an input loop that translates from English into French, allowing you to speak English to your computer and then have the computer translate your words into French. If you wish to test this out, you need to initiate this input loop by speaking.

/*
 * Creates the Active Event that translates from English speech to French.
 */
create-event:english.translate2french

  /*
   * Listens for speech input, and evaluates [onfinish] when speech has been captured.
   */
  magic-menu.listen
    onfinish

      /*
       * Checks to see if user spoke "thank you", at which our input loop is exited.
       */
      to-lower:x:/../*/text?value
      if:x:/-?value
        =:thank you
        magic-menu.listen:[Talk to me]
      else

        /*
         * Translates input to French
         */
        micro.google.translate:x:/../*/text?value
          dest-lang:fr-CA

        /*
         * Speeks French translation, and recursively invokes [english.translate2french] again.
         */
        magic-menu.speak:x:/-?value
          voice:fr-CA
          onfinish
            english.translate2french

The idea is that your phrases raise an "event" that executes some piece of code, facilitating you to literally create computer code by speaking in natural language to your computer. Below is a screenshot of how the GUI looks.

Image title

Consuming AI in byte sized applications is the best way to transform digitally. #BuiltOnAI, EdgeVerve’s business application, provides you with everything you need to plug & play AI into your enterprise.  Learn more.

Topics:
speech recognition ,ai ,hyperlambda ,nlp

Opinions expressed by DZone contributors are their own.

{{ parent.title || parent.header.title}}

{{ parent.tldr }}

{{ parent.urlSource.name }}