The company plans to make it possible to perform most functions through voice commands and expand the range of cellphones that can be operated this way.
The Quick Voice Activation function on NEC N-01C model that debuted in November works as follows:
Handset: "State a function you wish to use."
User: "Have any e-mail messages arrived?"
Handset: "I will check your e-mail."
The result, "two e-mail messages," is displayed on the screen.
Of the 28 cellphones that DoCoMo will introduce by next spring, 13 models, made by NEC, Sharp Corp., Panasonic Corp. and Fujitsu Ltd., will be equipped with this function.
The company expects that voice-activation cellphones will be popular with older people or disabled people who have difficulty using the keyboard, as well as by others when they are using both hands for cooking or other tasks.
When the user says, "I want to call A," A's phone number is displayed. Saying "alarm" brings up the alarm-setting display, while saying "calculator" brings up the calculator display.
After the display appears on the screen, the user must type in commands on the keyboard. The Quick Voice Activation function has to be set up in advance.
FueTrek Co., a start-up company in Osaka in which DoCoMo has an equity stake, developed the Quick Voice Activation function by combining the voice-recognition technology of the National Institute of Information and Communications Technology, the dialogue control technology of the Toyohashi University of Technology and the voice synthesis technology of the Nagoya Institute of Technology.
The function recognizes the user's voice by verifying it against a database of voices and words in the handset.
If the user utters irrevelant sounds like "um" at the beginning, the speech function cannot recognize the user, but the rate of recognition exceeds 90 percent as it gets more accustomed to the user, according to DoCoMo.
Researchers at the Toyohashi University of Technology say it is possible to perform interactions approaching human conversation using the XISL computer language for dialogue control.
Voice synthesis technology developed by the Nagoya Institute of Technology creates new voices from limited voice data.
Researchers at the university say the technology is easy to incorporate into cellphones because it is not necessary to store a large amount of data in the handset.