WEARABLE DEVICE WITH BLOCKED SENSOR DETECTION
Techniques, including devices and systems implementing the techniques, for using speech enhancement to provide optimal denoised output. One example system generally includes a wearable device, a first sensor coupled to the wearable device, a second sensor coupled to the wearable device, and one or more processors coupled to the wearable device. The one or more processors, individually or collectively, may be generally configured to receive, at the first sensor, a first audio signal, receive, at the second sensor, a second audio signal, and determine a condition of the wearable device based, at least in part, on (i) a sum energy comprising an energy of the first audio signal and an energy of the second audio signal in a frequency range and on (ii) an energy difference between the energy of the first audio signal and the energy of the second audio signal in the frequency range.
1 . A system comprising:
a wearable device;
a first sensor coupled to the wearable device;
a second sensor coupled to the wearable device; and
one or more processors coupled to the wearable device, the one or more processors, individually or collectively, being configured to:
receive, at the first sensor, a first audio signal;
receive, at the second sensor, a second audio signal; and
determine a condition of the wearable device based, at least in part, on a sum energy comprising an energy of the first audio signal and an energy of the second audio signal in a frequency range and on an energy difference between the energy of the first audio signal and the energy of the second audio signal in the frequency range.
2 . The system of claim 1 , wherein the one or more processors, individually or collectively, are further configured to:
determine a gain difference between the energy of the first audio signal and the energy of the second audio signal in the frequency range; and
apply a gain to the first audio signal to effectively compensate for the gain difference and form a scaled first audio signal.
3 . The system of claim 2 , wherein the one or more processors, individually or collectively, are further configured to:
determine an energy of the scaled first audio signal in the frequency range;
determine a scaled sum energy that comprises the energy of the scaled first audio signal and the energy of the second audio signal; and
determine a scaled energy difference between the scaled first audio signal and the second audio signal.
4 . The system of claim 3 , wherein the one or more processors, individually or collectively, are configured to determine the condition of the wearable device by:
determining a first ratio using the sum energy and the energy difference;
determining a second ratio using the scaled sum energy and the scaled energy difference;
determining a third ratio using the first ratio and the second ratio; and
determining that the condition of the wearable device is blocked when at least one of:
the gain is above a gain threshold,
the scaled energy difference is less than the energy difference by an energy threshold, or
the third ratio is greater than a ratio threshold.
5 . The system of claim 4 , wherein the one or more processors, individually or collectively, are further configured to:
determine the energy of the first audio signal in the frequency range; and
determine the energy of the second audio signal in the frequency range.
6 . The system of claim 5 , wherein the frequency range comprises 400 Hz to 1 kHz.
7 . The system of claim 4 , wherein when the condition of the wearable device is blocked, a condition of the second sensor is blocked when the gain is less than one, a condition of the first sensor is blocked when the gain is greater than one.
8 . The system of claim 7 , wherein when the condition of the wearable device is blocked, the one or more processors, individually or collectively, are further configured to:
when the condition of the second sensor is blocked, mix the first audio signal to form an output audio signal; and
when the condition of the first sensor is blocked, mix the second audio signal to form the output audio signal.
9 . The system of claim 1 , wherein a noise of the first audio signal and a noise of the second audio signal are both above a noise threshold.
10 . A method for audio signal processing in a wearable device, the method comprising:
receiving, at a first sensor included in the wearable device, a first audio signal;
receiving, at a second sensor included in the wearable device, a second audio signal; and
determining a condition of the wearable device based, at least in part, on a sum energy comprising an energy of the first audio signal and an energy of the second audio signal in a frequency range and on an energy difference between the energy of the first audio signal and the energy of the second audio signal in the frequency range.
11 . The method of claim 10 , further comprising:
determining a gain difference between the energy of the first audio signal and the energy of the second audio signal in the frequency range; and
applying a gain to the first audio signal to effectively compensate for the gain difference and form a scaled first audio signal.
12 . The method of claim 11 , further comprising:
determining an energy of the scaled first audio signal in the frequency range;
determining a scaled sum energy that comprises the energy of the scaled first audio signal and the energy of the second audio signal; and
determining a scaled energy difference between the scaled first audio signal and the second audio signal.
13 . The method of claim 12 , wherein determining the condition of the wearable device comprises:
determining a first ratio using the sum energy and the energy difference;
determining a second ratio using the scaled sum energy and the scaled energy difference;
determining a third ratio using the first ratio and the second ratio; and
determining that the condition of the wearable device is blocked when at least one of:
the gain is above a gain threshold,
the scaled energy difference is less than the energy difference by an energy threshold, or
the third ratio is greater than a ratio threshold.
14 . The method of claim 13 , wherein the frequency range comprises 400 Hz to 1 kHz.
15 . The method of claim 13 , wherein when the condition of the wearable device is blocked, a condition of the second sensor is blocked when the gain is less than one, a condition of the first sensor is blocked when the gain is greater than one.
16 . A non-transitory computer-readable medium comprising computer-executable instructions that, when executed by one or more processors of a wearable device, cause the wearable device to perform a method for audio signal processing, the method comprising:
receiving, at a first sensor included in the wearable device, a first audio signal;
receiving, at a second sensor included in the wearable device, a second audio signal; and
determining a condition of the wearable device based, at least in part, on a sum energy comprising an energy of the first audio signal and an energy of the second audio signal in a frequency range and on an energy difference between the energy of the first audio signal and the energy of the second audio signal in the frequency range.
17 . The non-transitory computer-readable medium of claim 16 , wherein the method further comprises:
determining a gain difference between the energy of the first audio signal and the energy of the second audio signal in the frequency range; and
applying a gain to the first audio signal to effectively compensate for the gain difference and form a scaled first audio signal.
18 . The non-transitory computer-readable medium of claim 17 , wherein the method further comprises:
determining an energy of the scaled first audio signal in the frequency range;
determining a scaled sum energy that comprises the energy of the scaled first audio signal and the energy of the second audio signal; and
determining a scaled energy difference between the scaled first audio signal and the second audio signal.
19 . The non-transitory computer-readable medium of claim 18 , wherein determining a condition of the wearable device comprises:
determining a first ratio using the sum energy and the energy difference;
determining a second ratio using the scaled sum energy and the scaled energy difference;
determining a third ratio using the first ratio and the second ratio; and
determining that the condition of the wearable device is blocked when at least one of:
the gain is above a gain threshold,
the scaled energy difference is less than the energy difference by an energy threshold, or
the third ratio is greater than a ratio threshold.
20 . The non-transitory computer-readable medium of claim 19 , wherein the frequency range comprises 400 Hz to 1 kHz.