Speech recognition is a computer-processing technology that converts the human voice into text or commands, making possible more user-friendly products by enabling equipment to be controlled and information to be input into computer systems by voice alone.

Canon's VoiceMaster speech-recognition engine achieves a recognition rate among the industry's highest, with a high level of accuracy even when the speaker's voice has not been registered beforehand. The engine is capable of quickly recognizing not only individual words, but also continuous speech such as phrases and sentences, and can operate even amid surrounding noise, thanks to Canon's environment-response technology. The system handles both English and Japanese, and the speech-recognition engine lineup offers recognition vocabularies ranging from one hundred to tens of thousands of words, based on the intended use and performance capabilities of the target device. With the aim of improving product usability, Canon is currently working to develop and expand this technology, moving ahead in the research and development of next-generation speech-recognition technologies and engines to provide such functions as multi-language support.

* VoiceMaster is a Japanese trademark.

Speech Recognition Flow