The next generation of voice-based interface technology will enable easy- to-use automation of new and existing communication services, making human-machine interaction more natural.
Loyd V Allen · 2002
ngineers have long sought to design systemsthat understand and act upon spoken lan-guage. Extracting meaning from natural,unconstrained speech over the telephone istechnically challenging, and quantifyingsemantic content is crucial for engineering and eval-uating such systems.Traditional menu-driven speech recognition sys-tems force users to learn the machine’s jargon, butmany people are unwilling or unable to navigate suchhighly structured interactions. AT&T’s “How MayI Help You?”(HMIHY) technology (http://www.research.att.com/~algor/hmihy) shifts the burden tothe machine by requiring it to adapt to human lan-guage and understand what people actually sayrather than what a system designer expects them to say. The intuition underlying our approach is that fora given task, some linguistic events are more