Information processing system and information processing method
AssigneeSony Corporation
InventorsShinichi Kawano, Yuhei Taki
Life of this patent
Estimating the term…
Abstract
There is provided an information processing system enabling a user to provide easily an instruction on whether to continue speech recognition processing on sound information, the information processing system including: a recognition control portion configured to control a speech recognition portion so that the speech recognition portion performs speech recognition processing on sound information input from a sound collection portion. The recognition control portion controls whether to continue the speech recognition processing on the basis of a gesture of a user detected at predetermined timing.
Claims15
An information processing system comprising:
- a recognition control portion configured to
- control a speech recognition portion so that the speech recognition portion performs speech recognition processing on sound information input from a sound collection portion, and
- detect a duration, in which a volume of the sound information continuously falls below a reference volume, reaching a predetermined target time after the speech recognition processing has started;
- and
- an output control portion configured to cause an output portion to output a motion object that is displayed by a display after the detected duration,
- wherein the recognition control portion controls whether to continue the speech recognition processing, based on a degree of coincidence between a trajectory of a viewpoint of a user detected after the output portion outputs the motion object and a trajectory of the motion object displayed by the display,
- wherein the recognition control portion and the output control portion are each implemented via at least one processor.
An information processing method comprising:
- performing speech recognition processing on sound information input from a sound collection portion;
- detecting a duration, in which a volume of the sound information continuously falls below a reference volume, reaching a predetermined target time after the speech recognition processing has started;
- outputting a motion object that is displayed by a display after the detected duration;
- and
- controlling, by a processor, whether to continue the speech recognition processing, based on a degree of coincidence between a trajectory of a viewpoint of a user detected after the outputting of the motion object and a trajectory of the motion object displayed by the display.
A non-transitory computer-readable medium having embodied thereon a program, which when executed by a computer causes the computer to execute a method, the method comprising:
- performing speech recognition processing on sound information input from a sound collection portion;
- detecting a duration, in which a volume of the sound information continuously falls below a reference volume, reaching a predetermined target time after the speech recognition processing has started;
- outputting a motion object that is displayed by a display after the detected duration;
- and
- controlling, by a processor, whether to continue the speech recognition processing, based on a degree of coincidence between a trajectory of a viewpoint of a user detected after the outputting of the motion object and a trajectory of the motion object displayed by the display.
Dependencies are inferred from the claim text.