Method and apparatus for enabling multimodal tags in a communication device by discarding redundant information in the tags training signals

1. A method comprising:

receiving, by an electronic device, one or more first user inputs indicative of a first selection of a first input modality for a multimodal tag and a second input modality for the multimodal tag, wherein the first and second input modalities are selected from a set of input modalities comprising at least three different types of input modalities;

detecting, with the first input modality of the electronic device, a first training signal;

detecting, with the second input modality of the electronic device, a second training signal;

creating a multimodal tag by at least discarding redundant or non-discriminative information associated with each of the first and second training signals and producing a training signal feature sequence that defines one or more salient features of both the first and the second training signals as a function of time;

receiving at least one second user input indicative of a second selection of a function from a plurality of selectable functions; and

associating the function with the multimodal tag.

2. The method of claim 1 , further comprising:

receiving a plurality of input signals; and

in response to determining that the plurality of input signals correspond to the multimodal tag performing the function associated with the multimodal tag.

3. The method of claim 2 , wherein at least one of the plurality of input signals includes a tactile sensor based signal.

4. The method of claim 2 , wherein determining whether the received plurality of input signals correspond to the multimodal tag is done using a pattern matching technique.

5. The method of claim 4 , wherein the pattern matching technique is performed on a first feature sequence and a second feature sequence, wherein the first feature sequence characterizes the first input signal as a function of time and the second feature sequence characterizes the second input signal as a function of time.

6. The method of claim 1 , wherein the second training signal is a motion signal detected by a motion and the first training signal is an audio signal detected by an audio sensor.

7. The method of claim 6 , further comprising: detecting the first training signal by at least analyzing the audio signal using speech analysis techniques.

8. The method of claim 1 , wherein the second training signal is tactile signal detected by a tactile sensor and the first training signal is an audio signal detected by an audio sensor.

9. The method of claim 8 , wherein the tactile signal is one of a touch based input, a key press input, or a stylus input.

10. The method of claim 1 , further comprising detecting at least a third training signal, wherein creating the multimodal tag comprises combining the first, second, and third training signals into the training signal feature sequence.

11. The method of claim 1 , wherein registering the training signal feature sequence comprises storing the training signal feature sequence as the multimodal tag.

12. The method of claim 1 , wherein the second training signal is a motion based signal detected by a motion sensor and the first training signal is a tactile signal detected by a tactile sensor.

13. The method of claim 1 , wherein at least one of the first training signal and the second training signal is based on at least one of velocity, acceleration, direction, trajectory, gravity, or rotation detected by a motion sensor.

14. The method of claim 1 , wherein the second training signal is a motion based signal and the first training signal is a gesture based signal.

15. The method of claim 1 , wherein the second training signal is a gesture based signal and the first training signal is a tactile sensor based signal.

16. The method of claim 1 , wherein creating the multimodal tag further comprises combining the first and second training signals into the training signal feature sequence based on a temporal order at which the first training signal and the second training signal are detected.

17. The method of claim 1 , wherein the first input signal is a motion based signal indicative of a shaking motion, the second input signal is a tactile based signal indicative of a key press, and the function is dialing a an emergency response service.

18. An electronic device comprising:

a first sensor configured as a first input modality for detecting a first training signal associated with a multimodal tag;

a second sensor configured as a second input modality for detecting a second training signal associated with a multimodal tag;

a third sensor configured as a third input modality for detecting a third training signal associated with a multimodal tag;

a memory for storing registered multimodal tags and associated functions; and

a processor operably coupled to the first sensor, the second sensor, and the third sensor, wherein the processor is configured to:

receive one or more first user inputs indicative of a first selection of the first input modality, the second input modality, and the third input modality;

receive the first training signal, the second training signal, and the third training signal;

receive at least one second user input indicative of a second selection of a function from a plurality of selectable functions;

discard redundant or non-discriminative information associated with each of the first, second, and third training signals;

produce a training signal feature sequence that defines one or more salient features of a combination of the first, second, and third training signals as a function of time; and

register the training signal feature sequence as a multimodal tag by at least storing the user selected function in association with the multimodal tag.

19. The electronic device of claim 18 , wherein producing the training signal feature sequence comprises combining the first, and second training signals into the training signal feature sequence based on a temporal order at which the first, and second training signals are received.

20. The electronic device of claim 18 , wherein one of the first and second sensors are motion based sensors for detecting motion of the electronic device and the third sensor is a tactile based sensor for detecting tactile input associated with the electronic device.

21. A method comprising:

receiving, by an electronic device, one or more first user inputs indicative of:

a first selection of at least a first input modality for a multimodal tag,

a second input modality for the multimodal tag, and

a third input modality for the multimodal tag;

detecting, with the first input modality of the electronic device, a first training signal;

detecting, with the second input modality of the electronic device, a second training signal;

detecting, with the third input modality of the electronic device, a third training signal;

discarding redundant or non-discriminative information associated with each of the first, second, and third training signals;

creating a multimodal tag by at least combining the first, second, and third training signals into one or more training signal feature sequences that define one or more salient features of a combination of the first, second, and third training signals as a function of time;

receiving at least one second user input indicative of a second selection of a function from a plurality of selectable functions;

and associating the function with the multimodal tag.