This technology, "Hyper Speech Markup Language", seamlessly integrates speech enabled web content with existing visual pages. This technology improves web accessibility from hands-free mobile devices to assistive technology through extending voice commands beyond conventional speech-to-text and text-to-speech approaches. Speech- and audio-based user interfaces developed with this technology invite users to speak about page content, with common-language phrases and questions in place of required keywords or specific link titles. Voice-user interfaces (VUIs) developed with this technology complement visual webpages. They are dynamically generated on each page, and tailor the speech recognition and speech synthesis to the current page's unique content. The audio component can be invisible to non-voice-command users, and web developers can choose the level of speech-enabled design to suit their needs.
A speech recognition grammar is embedded in webpages using this technology's "hyper-speech" links. But Automated Speech Recognition (ASR) programming is not required to build flexible, sophisticated websites using this technology's audio and speech integration. It requires essentially no bandwidth, because binary code is not transferred between the browser and the server. And it is compatible with popular browsers -- no additional software or plugins required! This scheme represents a breakthrough in assistive technology, in addition to providing an additional interactive dimension for small screens and hands-free applications. Speech enabled website design, with this technology, can provide an equally vibrant user experience for users with impaired vision or dexterity.
Michael Charney
Tech Ventures Reference: IR M01-011, IR 1489