Skip to main content

IBM Cambridge Research Center

  Project: Lotus Conversational Interface

Researchers: Steven Ross, Julie Alweis, Robert Armes, Beth Brownholtz

A Cambridge Speech Initiative Group Project

The Cambridge Speech Initiative Group has been focusing on applying speech recognition and speech synthesis technology to create a conversational, cross-application, speech-enabled environment.

The project emphasizes natural language queries, particularly ones where users speak to background applications. While maintaining focus on their primary task, users will be able to make requests such as "Schedule a meeting with John on Tuesday" or "What is the current IBM stock price?" and receive synthesized speech responses. The conversational agent can ask clarifying questions, and the user will be able to participate in the conversation without changing focus or bringing hidden application windows to the foreground.

The group is initially working on speech enabling desktop applications such as email, calendar, and instant-messaging. In the longer term, our technology has potential to be applied to other areas such as knowledge management, as well as venues such as telephony and ubiquitous computing devices. The goal is to identify application "sweet spots" -- places where current speech recognition and synthesis technology can match user needs and expectations particularly well -- that will suggest product opportunities that can exploit IBM's investment in speech technology.