PC Magazine reports that ViaVoice 4.4 supports "free-form commands," including command synonyms and flexible ordering of command parameters. Previous voice technology required users to stick to fairly limited scripts when issuing voice commands.
The
PC Magazine article touts the ViaVoice 4.4 release as a major advance, and a landmark on IBM's five-year goal of achieving "super-human" speech recognition; that is, computer speech recognition that equals and even surpasses human capabilities.
Also touted in the new release are improved acoustic models, which help sift the wheat of voice from the chaff of background noise, as well as better filtering of transient noises, such as honking horns blown at drivers distracted by conversations with their gadgets.
Still, IBM claims that the more natural conversational patterns permitted by ViaVoice 4.4 are less distracting to drivers than trying to recall more rigid command syntaxes. Jim Holland, product line manager of embedded speech at IBM, stated, "Speech recognition has gotten a bad rap in recent years, but it has quietly been maturing into a truly useful technology. The ability to use natural language commands allows the device to become a more instinctive part of a user's daily routine."
Typical applications for the new ViaVoice release include car radio tuning, phone dialing, and on-the-fly translations of multimedia content, for example to create subtitles for foreign films,
PC Magazine says.
In a related story,
eWEEK has written about a Wake Forest University project that aims to combine IBM's voice recognition technology with web services, allowing students with cell phones to perform such feats as browsing class schedules or the library card catalog, or checking washing machine availability.
The
PC Magazine article can be found
here, and the
eWEEK article is available
here.
Related Stories:
