Implementing the Speechly Voice UI for the Web – VIDEO

The web is still the largest digital surface area around. It is at least two orders of magnitude larger than mobile today and many more orders of magnitude larger than smart speakers. And yet, there is close to zero conversation about applying a voice user interface to the web. This is especially puzzling since so many consumers and professionals have been at home over the past year and using their laptops and desktops with higher frequency than when they were out of the house more regularly where mobile is often the only convenient option for digital connectivity.

Web UI has changed little in the last decade from a functional standpoint. However, in the video below, Speechly lays out specific use cases a voice UI is far superior to the point, click, and type interface that dominates the web. This is not a turn-based voice assistant conversation. Speechly has introduced a different approach that is multimodal by design, doesn’t attempt to engage in a conversation, and instead focuses on completing the user objective as quickly as possible using voice as an input and the screen as an output.

How to Deploy a Voice UI for the Web

In today’s webinar, the featured speakers Hannes Heikinheimo, Speechly co-founder and CTO, and Antti Ukkonen, Speechly head of product and NLU, show end-user demonstrations of different voice use cases on the web, discuss Speechly’s technical approach, and walk through how to implement the solution with just two lines of code and some automated training. There is also an announcement about Speechly’s forthcoming integration with Wix towards the end.


Watch the full video or follow the time-stamps below to see the demonstration, reviews of the technology stack, and developer tools in today’s webinar.

  • 2:32: Bret Kinsella on the web UI model and the opportunity for voice AI
  • 4:01: Introduction of the featured speakers, Hannes Heikenheimo and Antii Ukkonen
  • 5:17: The fastest voice UI for the Web
  • 6:03: End-user demos
  • 10:17: Difference between the fast voice UI control model and the turn-based voice assistants
  • 11:58: Speechly programming model
  • 19:55: Demo of how Speechly is deployed in the web
  • 25:55: Speechly implementation through WIX announcement

Speechly Shows a New Model for Voice UI on the Web – VIDEO

Finnish Voice Recognition Startup Speechly Closes €2M Funding Round

Val Kilmer Gets a New Synthetic Voice Replica from Sonantic