Smart speaker device with cognitive sound analysis and response

A computer program product comprising a computer readable storage medium having a computer readable program stored therein, wherein the computer readable program, when executed in a smart speaker device, causes the smart speaker device to:

• capture one or more audio samples from a monitored environment;
• store the captured one or more audio samples in an audio sample buffer, wherein the audio sample buffer stores an amount of captured audio samples corresponding to a specified time window for detecting events occurring within the monitored environment;
• analyze the one or more captured audio samples to determine whether the one or more captured audio samples represent a potential wake sound for initiating a responsive action by the smart speaker device, wherein the potential wake sound is a variable sound determined to not be a normal ambient sound of the monitored environment;
• and
• in response to determining that the one or more captured audio samples represent a potential wake sound for initiating a responsive action by the smart speaker device, initiate performance of the responsive action by the smart speaker device, wherein determining whether the one or more captured audio samples represent a potential wake sound for initiating a responsive action by the smart speaker device, further comprises:
• learning privacy concern information of a user associated with the monitored environment;
• and
• applying a privacy firewall based on the privacy concern information of the user to the one or more captured audio samples to filter and disregard storing of captured audio samples in the audio sample buffer that are determined to be private based on the privacy concern information of the privacy firewall, wherein learning privacy concern information of the user comprises analyzing user cultural or social context associated with the user, determining a user personality type, and measuring user historical speaking tone of the user, and determining a privacy type or level of privacy associated with the user based on the learned privacy concern information, wherein applying the privacy firewall comprises filtering and disregarding storing of captured audio samples that correspond to the determined privacy type or level of privacy.

A smart speaker device comprising:

• a processor;
• a memory coupled to the processor;
• an audio capture device coupled to the processor and the memory;
• and
• an audio sample buffer coupled to the audio capture device, wherein the memory comprises instructions which, when executed by the processor, cause the processor to:
• capture, via the audio capture device, one or more audio samples from a monitored environment;
• store the captured one or more audio samples in the audio sample buffer, wherein the audio sample buffer stores an amount of captured audio samples corresponding to a specified time window for detecting events occurring within the monitored environment;
• analyze the one or more captured audio samples to determine whether the one or more captured audio samples represent a potential wake sound for initiating a responsive action by the smart speaker device, wherein the potential wake sound is a variable sound determined to not be a normal ambient sound of the monitored environment;
• and
• in response to determining that the one or more captured audio samples represent a potential wake sound for initiating a responsive action by the smart speaker device, initiate performance of the responsive action by the smart speaker device, wherein determining whether the one or more captured audio samples represent a potential wake sound for initiating a responsive action by the smart speaker device, further comprises:
• learning privacy concern information of a user associated with the monitored environment;
• and
• applying a privacy firewall based on the privacy concern information of the user to the one or more captured audio samples to filter and disregard storing of captured audio samples in the audio sample buffer that are determined to be private based on the privacy concern information of the privacy firewall, wherein learning privacy concern information of the user comprises analyzing user cultural or social context associated with the user, determining a user personality type, and measuring user historical speaking tone of the user, and determining a privacy type or level of privacy associated with the user based on the learned privacy concern information, wherein applying the privacy firewall comprises filtering and disregarding storing of captured audio samples that correspond to the determined privacy type or level of privacy.