Texting and reading a web-page are not comfortable paradigms for accessing information, especially when mobile. The Verbal Internet changes this radically. Using the heavy compute power that will become available via the ‘cloud,’ a smart phone can couple a speaker to voice recognition software, interpret queries or actions, and respond verbally. The technology is mostly available, though integration is not.

Now, this is not just reading a page of Google results out loud. The pain level would be way too high! The Verbal Internet (VI) requires that both queries and response be concise and useful, and that navigation through unwanted junk be easy. This means a different type of web page. As an example, when I ask Google “What is the meaning of life?” I expect “42” as the answer, not 1.6 million references to hitchhikers.

Logistically, a VI webpage will be primed to take voice input and send it to a recognition application. The page will provide context vocabulary tables (since limited vocab recognition works best) and will use the same scripting and Java approach to parse the recognized input when returned from the app.

The recognition apps will run in a very-dense virtual machine context, so cost per service will be suitably low. Obviously, with location sensing, such uses as GPS, tour guides, and how-to instructions can be moved to verbal contexts on your phone.

The cost is rethinking how web page iterations occur. Going back to Google, it’s clear this is a major step.

The extension of Internet addressing via IPV6 will allow the verbal interface to become ubiquitous. Verbal TV remote controls and light switches will be achievable, advertising will change and as sophistication improves, everything will be personalized to much greater levels than today.