Modification of voice commands based on sensitivity

1. A computer-implemented method for modifying an input command of a virtual assistant, comprising:

receiving the input command from a user;

determining a contextual environment surrounding the virtual assistant;

detecting, based on the contextual environment, sensitive information in the input command;

replacing the sensitive information with generic information; and

responding to the input command with an output response based, in part, on the generic information.

2. The computer-implemented method of claim 1 , wherein determining the contextual environment surrounding the virtual assistant is based in part on at least one operation selected from the group consisting of:

identifying a second user present in the contextual environment;

identifying a location of the contextual environment;

identifying spoken content between the user and the second user present in the contextual environment; and

identifying an emotional state of one or more users present in the contextual environment.

3. The computer-implemented method of claim 1 , wherein detecting the sensitive information in the input command comprises:

segmenting the input command into a plurality of portions;

analyzing, using natural language processing, each portion with respect to the contextual environment and one or more sensitivity rules; and

identifying one or more portions that meet the one or more sensitivity rules.

4. The computer-implemented method of claim 3 , wherein the one or more sensitivity rules are based on historical interaction data of the user related to the virtual assistant.

5. The computer-implemented method of claim 3 , wherein the one or more sensitivity rules are based on historical crowdsourced sensitivity data.

6. The computer-implemented method of claim 3 , wherein the one or more sensitivity rules are based on third-party marked sensitivity data.

7. The computer-implemented method of claim 3 , wherein the one or more sensitivity rules are weighted based on attributes related to the contextual environment.

8. A system comprising:

a processor; and

a computer-readable storage medium communicatively coupled to the processor and storing program instructions which, when executed by the processor, cause the processor to perform a method comprising:

receiving an input command from a user;

determining a contextual environment surrounding a virtual assistant;

detecting, based on the contextual environment, sensitive information in the input command;

replacing the sensitive information with generic information; and

responding to the input command with an output response based, in part, on the generic information.

9. The system of claim 8 , wherein determining the contextual environment surrounding the virtual assistant is based in part on at least one operation selected from the group consisting of:

identifying a second user present in the contextual environment;

identifying a location of the contextual environment;

identifying spoken content between the user and the second user present in the contextual environment; and

identifying an emotional state of one or more users present in the contextual environment.

10. The system of claim 8 , wherein detecting the sensitive information in the input command, comprises:

segmenting the input command into a plurality of portions;

analyzing, using natural language processing, each portion with respect to the contextual environment and one or more sensitivity rules; and

identifying one or more portions that meet the one or more sensitivity rules.

11. The system of claim 10 , wherein the one or more sensitivity rules are based on historical interaction data of the user related to the virtual assistant.

12. The system of claim 10 , wherein the one or more sensitivity rules are based on historical crowdsourced sensitivity data.

13. The system of claim 10 , wherein the one or more sensitivity rules are based on third-party marked sensitivity data.

14. The system of claim 10 , wherein the one or more sensitivity rules are weighted based on attributes related to the contextual environment.

15. A computer program product comprising a computer-readable storage medium having program instructions embodied therewith, the program instructions executable by a processor to cause the processor to perform a method comprising:

receiving an input command from a user;

determining a contextual environment surrounding a virtual assistant;

detecting, based on the contextual environment, sensitive information in the input command;

replacing the sensitive information with generic information; and

responding to the input command with an output response based, in part, on the generic information.

16. The computer program product of claim 15 , wherein determining the contextual environment surrounding the virtual assistant is based in part on at least one operation selected from the group consisting of:

identifying a second user present in the contextual environment;

identifying a location of the contextual environment;

identifying spoken content between the user and the second user present in the contextual environment; and

identifying an emotional state of one or more users present in the contextual environment.

17. The computer program product of claim 15 , wherein detecting the sensitive information in the input command, comprises:

segmenting the input command into a plurality of portions;

analyzing, using natural language processing, each portion with respect to the contextual environment and one or more sensitivity rules; and

identifying one or more portions that meet the one or more sensitivity rules.

18. The computer program product of claim 17 , wherein the one or more sensitivity rules are based on historical interaction data of the user related to the virtual assistant.

19. The computer program product of claim 17 , wherein the one or more sensitivity rules are based on historical crowdsourced sensitivity data.

20. A computer-implemented method for modifying an input command of a virtual assistant in the presence of other users, comprising:

monitoring, by the virtual assistant, one or more Internet of Things (IoT) data feeds from one or more IoT devices;

receiving, by the virtual assistant, the input command from a user;

detecting, by the virtual assistant, sensitive information in the input command;

determining, by the virtual assistant, if one or more other users are present in an environment surrounding the virtual assistant;

removing, by the virtual assistant and in response to determining that one or more other users are present in the environment, the sensitive information from the input command and any associated sensitive information from an output response; and

responding, by the virtual assistant, with the output response.

21. The computer-implemented method of claim 20 , wherein the one or more IoT devices are selected from the group consisting of: a camera; a smart speaker;

a sensor; and a smart phone.

22. The computer-implemented method of claim 20 , wherein detecting the sensitive information in the input command, comprises:

segmenting, by the virtual assistant, the input command into a plurality of portions;

analyzing, by the virtual assistant and using natural language processing, each portion with respect to a contextual environment and one or more sensitivity rules; and

identifying, by the virtual assistant, one or more portions that meet the one or more sensitivity rules.

23. A computer-implemented method for requesting an alternative input method for receiving an input command by a virtual assistant, comprising:

monitoring, by the virtual assistant, one or more Internet of Things (IoT) data feeds from one or more IoT devices;

receiving, by the virtual assistant, the input command from a user;

detecting, by the virtual assistant, sensitive information in the input command;

determining, by the virtual assistant, if a second user is present in an environment surrounding the virtual assistant;

determining, by the virtual assistant and in response to determining that the second user is present in the environment, if an output response can be generated without the sensitive information; and

requesting, by the virtual assistant and in response to not being able to generate the output response without the sensitive information, that the user restate the input command using an alternative input method.

24. The computer-implemented method of claim 23 , wherein the one or more IoT devices are selected from the group consisting of: a camera; a smart speaker; a sensor; and a smart phone.

25. The computer-implemented method of claim 23 , further comprising:

receiving, by the virtual assistant, a second input command from a second IoT device; and

responding, by the virtual assistant, to the second IoT device with an output response.