Voice activated device
A voice activated camera is described which allows users to take remote photographs by speaking one or more keywords. In a preferred embodiment, a speech processing unit is provided which is arranged to detect extended periodic signals from a microphone of the camera. A control unit is also provided to control the taking of a photograph when such an extended periodic component is detected by the speech processing unit.
1. An apparatus for controlling a device, the apparatus comprising:
a receiver operable to receive a speech signal;
a divider operable to divide the speech signal into a sequence of speech frames, each speech frame representing a time portion of the speech signal;
a processor operable to determine an auto-correlation function for each speech frame;
a peak determiner operable to locate a maximum auto-correlation value within a predetermined part of said auto-correlation function for each speech frame;
a periodicity measurer operable to determine a periodicity measure of each speech frame by using the maximum auto-correlation value of each speech frame;
a variability measurer operable to determine, from periodicity measures of a plurality of consecutive speech frames, a plurality of variability measures by using different calculation methods;
a detector operable to combine the plurality of variability measures and to detect an extended periodic portion from the combined variability measures; and
a controller operable to control said device in dependence upon a detection made by said detector.
2. An apparatus according to claim 1 , wherein said detector is operable to linearly combine said plurality of variability measures.
3. An apparatus according to claim 1 , wherein said detector is operable to combine said plurality of variability measures in a non-linear manner.
4. An apparatus according to claim 1 , wherein said detector is operable to compare the combined periodicity measures with a threshold value.
5. An apparatus according to claim 1 , wherein said predetermined part corresponds to an input frequency of the speech signal in the range of 100 Hertz to 300 Hertz.
6. An apparatus according to claim 1 , wherein said processor is operable to calculate:
v
i
=
A
i
(
L
MAX
)
A
i
(
0
)
where A(0) is the value of the auto-correlation function for a zero shift and A(L max ) is said maximum auto-correlation value determined by said peak determiner.
7. An apparatus according to claim 1 , further comprising an analogue to digital convener operable to sample the received speech signal.
8. An apparatus according to claim 7 , wherein said divider is operable to divide said speech signal so that each frame comprises the same number of samples.
9. An apparatus according to claim 1 , further comprising a microphone, wherein said receiver is operable to receive the speech signal from said microphone.
10. An apparatus according to claim 1 , wherein said device is an image capture device and wherein said controller is operable to activate the image capture device to capture an image.
11. An apparatus according to claim 1 , wherein said detector is operable to process the periodicity measures from successive speech frames corresponding to a predetermined duration of speech.
12. An apparatus according to claim 11 , wherein said detector is operable to process the periodicity measures from successive speech frames corresponding to a portion of speech having a duration between half a second and one and a half seconds.
13. A computer readable medium storing computer executable instructions for causing a computer to perform a controlling method for a device, the computer executable instructions comprising instructions for:
receiving a speech signal;
dividing the speech signal into a sequence of speech frames, each speech frame representing a time portion of the speech signal;
processing each speech frame to determine an auto-correlation function for each speech frame;
locating a maximum auto-correlation value within a predetermined part of the auto-correlation function for each speech frame;
determining a periodicity measure of each speech frame by using the maximum auto-correlation value of each speech frame;
determining, from periodicity measures of a plurality of consecutive speech frames, a plurality of variability measures by using different calculation methods;
combining the plurality of variability measures;
detecting an extended periodic portion from the combined variability measures and outputting a detection; and
controlling the device in accordance with the detection.
14. Computer executable instructions for controlling a computer to perform a controlling method for a device, the computer executable instructions comprising instructions for:
receiving a speech signal;
dividing the speech signal into a sequence of speech frames, each speech frame representing a time portion of the speech signal;
processing each speech frame to determine an auto-collation function for each speech frame;
locating a maximum auto-collation value within a predetermined part of the auto-correlation function for each speech frame;
determining a periodicity measure of each speech frame by using the maximum auto-correlation value of each speech frame;
determining, from periodicity measures of a plurality of consecutive speech frames, a plurality of variability measures by using different calculation methods;
combining the plurality of variability measures;
detecting an extended periodic portion from the combined variability measures and outputting a detection; and
controlling the device in dependence upon the detection.
15. An apparatus for controlling a device, the apparatus comprising:
means for receiving a speech signal;
means for dividing the speech signal into a sequence of speech frames, each speech frame representing a time portion of the speech signal;
means for determining an auto-correlation function for each speech frame;
means for locating a maximum auto-correlation value within a predetermined part of the auto-correlation function for each speech frame;
means for determining a periodicity measure of each speech frame by using the maximum auto-correlation value of each speech frame;
means for determining, from periodicity measures of a plurality of consecutive speech frames, a plurality of variability measures by using different calculation methods;
means for combining the plurality of variability measures;
means for detecting an extended periodic portion from the combined variability measures; and
means for controlling the device in accordance with a detection made by said means for detecting.