Location-based voice recognition system with voice command
AssigneeLuxrobo Co., Ltd.
InventorsSeungmin Baek, Seungbae Son
Life of this patent
Estimating the term…
Abstract
An object of the present invention is to facilitate recognition of a voice command of a user in a situation where multiple devices including microphones are connected through a sensor network. A relative location of each device is determined and a location and a direction of the user are tracked through a time difference in which the voice command is applied. The command is interpreted based on the location and the direction of the user. Such a method as a method for a sensor network, Machine to Machine (M2M), Machine Type Communication (MTC), and Internet of Things (IoT) may be used for an intelligent service (smart home, smart building, etc.), digital education, security and safety related services, and the like.
Claims21
A location-based voice recognition system comprising:
- a plurality of voice command receiving devices including at least one microphone;
- and
- a sensor network control device connected with the plurality of voice command receiving devices through a sensor network,
- wherein the sensor network control device includes
- a sensor location determining unit determining relative locations of the plurality of voice command receiving devices,
- a user location calculating unit calculating a location of a user based on the relative locations,
- an utterance direction calculating unit calculating a converted voice magnitude based on voice magnitudes of the plurality of respective voice command receiving devices and determining an utterance direction range of the user based on the converted voice magnitude, and
- a target device selecting unit selecting a target voice command receiving device as a target of a voice command among the plurality of voice command receiving devices according to the utterance direction range,
- wherein the converted voice magnitude is a voice magnitude when it is assumed that the plurality of voice command receiving devices is at the same distance from the user;
- wherein the target device learning unit performs machine learning on the voice command of the user by using the voice command history including the success or failure of the voice command which is determined by the user's location information, utterance direction, and whether to utter a continuous voice command of the same content in the same utterance direction during a specific time period or whether there is a negative word in the continuous command;
- wherein the target device determining unit determines the content of the voice command according to the probability of each voice command based on the user's speech direction, location information, and whether the voice command is successful based on the result of the machine learning.
A location-based voice recognition method in a location-based voice recognition system including a plurality of voice command receiving devices including at least one microphone and a sensor network control device connected with the plurality of voice command receiving devices through a sensor network, the method comprising:
- determining relative locations of the plurality of voice command receiving devices;
- calculating a location of a user based on the relative locations;
- calculating a converted voice magnitude based on voice magnitudes of the plurality of respective voice command receiving devices and determining an utterance direction range of the user based on the converted voice magnitude;
- and
- selecting a target voice command receiving device as a target of a voice command among the plurality of voice command receiving devices according to the utterance direction range,
- wherein selecting the target voice command receiving device includes:
- performing machine learning on the user's voice command by using the user's location information and a voice command history including the range of utterance direction and success or failure of the voice command;
- and
- determining the content of the voice command based on the user's speech direction, location information, and probability of each voice command based on the success or failure of the voice command based on the machine learning, and determining the target voice command receiving device based on the determined content of the voice command,
- wherein the converted voice magnitude is a voice magnitude when it is assumed that the plurality of voice command receiving devices is at the same distance from the user, and
- wherein the success or failure of the voice command is determined according to whether or not a continuous voice command having the same content is uttered in the same range of utterance direction for a specific time period or whether there is a negative word in the continuous command.
Dependencies are inferred from the claim text.