| 摘要 |
Voices are prevented from being recognized with poor accuracy when a speaker is not close to a sound pickup device. A speech recognition apparatus (10) includes a sound pickup device (16) picking up sounds, a photographing device (18) photographing images of a speaker making voices into the sound pickup device (16), a voice recognition function unit (132) recognizing voices on the basis of the picked-up voices, and a recognition/learning determination unit (142) restricting the voice recognition function unit (132) from recognizing voices when the photographed images do not contain speaker images showing at least part of the speaker.
|
