JSFoo 2017

JSFoo is a conference about JavaScript and everything related.


Voice on the web

Submitted by Vamsidhar Bethanabatla (@vamsib) on Thursday, 4 May 2017

Section: Full Talk Technical level: Intermediate Status: Rehearsal ongoing


Voice/Speech as a user interface medium is growing in popularity thanks to applications like Alexa, Cortana, Siri and Google. For long we have used a level of indirection to deal with machines. For e.g. using a mouse to click a button on screen or using arrow keys to move the content. Touch interfaces came and removed this level of inderection and allowed us to interact with content directly. Voice has the potential to bring us even closer to machines. With AI and intelligent agents growing, the interfaces of systems are not going to be limited enough to fit on a screen. These agents keep learning and expose new features. Voice/Speech is a fluid mechanism that can adapt to this ever growing interface. How can we build voice driven apps on the web? What standards are in place to guide these apps? Doesn’t this require massive infrastructure and ML algorithms to make it work? These are some questions we’ll explore and look at the simplest ways to get started and alternatives available. A lot can already be achieved using JavaScript and web standards.


  • Introduction - Why speech/voice as interface, examples of possible applications
  • Web speech API - Speech recognition (continuous and with interim results), speech synthesis, grammar
  • Open source libraries (CMU Sphinx)
  • A demo
  • Questions

Speaker bio

Vamsi has been a UI/Frontend developer for over 12 years. Currently working as a Lead Consultant, UI at ThoughtWorks. It was while working at Yahoo! that he was inspired to develop the best UIs which follow strict standards and are accessible. Still carrying that inspiration he is passionate about the quality of user interfaces and newer possibilities. One such possibility, the “Conversational UIs” was introduced to him at ThoughtWorks. For the past several months he has been exprimenting with web speech api, CMU sphinx and how these technologies can be brought together to develop meaningful web apps. One such attempt is to build an extensible voice assistant using open technologies.




  •   Sandhya Ramesh (@sandhyaramesh) 2 years ago

    Hi Vamsidhar, we’re evaluating proposals right now. Could you go ahead and submit your slide deck? Also, a quick recommendation would be to make your talk more hands-on, so more code and more demos would be great!

  •   Vamsidhar Bethanabatla (@vamsib) Proposer 2 years ago

    Hi Sandhya, yes there’s going to be code involved. A demo will also be there. I am Half way through the slide deck. Need couple of days more. Thanks!

  •   Vamsidhar Bethanabatla (@vamsib) Proposer 2 years ago

    Uploaded the slide deck.

    •   Sandhya Ramesh (@sandhyaramesh) 2 years ago


Login with Twitter or Google to leave a comment