Audio signals output
Examples are described herein for detecting, and outputting audio signals to, wearable audio output devices. In various examples, a device may include a time-of-flight (ToF) sensor and a processor. The processor may analyze ToF sensor data generated by the ToF sensor. Based on the analysis, the processor may detect a wearable audio output device worn by an individual. In response to detecting the wearable audio output device, the processor may output an audio signal to the wearable audio output device.
1 . A device comprising:
a time-of-flight (ToF) sensor; and
a processor to:
analyze ToF sensor data generated by the ToF sensor;
based on the analysis, detect a wearable audio output device worn by an individual, and
in response to detecting the wearable audio output device, output an audio signal to the wearable audio output device.
2 . The device of claim 1 , further comprising a housing, wherein the ToF sensor is integral with the housing.
3 . The device of claim 1 , further comprising an additional sensor that is distinct from the ToF sensor, wherein the processor is to:
analyze additional sensor data generated by the additional sensor; and
detect the wearable audio output device based on the additional sensor data and the ToF sensor data.
4 . The device of claim 3 , wherein the additional sensor comprises a magnetic field sensor and the additional sensor data represents magnetic field distortion caused by a magnet in the wearable audio output device.
5 . The device of claim 1 , wherein the ToF sensor comprises an ultrasonic ToF sensor.
6 . A non-transitory computer-readable medium comprising instructions that, in response to execution of the instructions by a processor of a computing device, cause the processor to:
operate a depth sensor of the computing device to acquire depth sensor data that includes a proximity distribution between an individual and the depth sensor;
analyze the depth sensor data to detect presence of a wearable audio output device worn by the individual; and
based on the detected presence, output an audio signal to the wearable audio output device worn by the individual.
7 . The non-transitory computer-readable medium of claim 6 , comprising instructions to process the depth sensor data generated by the depth sensor based on a machine learning model to generate output indicative of the detected wearable audio output device.
8 . The non-transitory computer-readable medium of claim 7 , wherein the machine learning model comprises a convolutional neural network (CNN).
9 . The non-transitory computer-readable medium of claim 8 , wherein the CNN comprises a first CNN, the output comprises first output, and the medium comprises instructions to:
process the depth sensor data based on a second CNN to generate second output indicative of a location of the individual's head; and
crop the depth sensor data based on the second output prior to processing the depth sensor data based on the first CNN.
10 . The non-transitory computer-readable medium of claim 6 , comprising instructions to:
operate a magnetic field sensor to acquire magnetic field sensor data; and
analyze the magnetic field sensor data to confirm the presence of the wearable audio output device detected based on the depth sensor data.
11 . The non-transitory computer-readable medium of claim 10 , comprising instructions that, in response to a determination that the magnetic field sensor data fails to confirm the depth sensor data, cause the processor to provide audio or visual output to the individual that solicits input from the individual, wherein the audio signal is output to the wearable audio output device based on the input received from the individual.
12 . A non-transitory computer-readable medium comprising instructions that, in response to execution of the instructions by a processor of a computing device, cause the processor to:
operate first and second contactless sensors to obtain, respectively, first and second sensor data;
analyze the first and second sensor data to detect headphones worn by an individual; and
based on the detected headphones, output an audio signal generated by the computing device to the headphones worn by the individual.
13 . The non-transitory computer-readable medium of claim 12 , wherein the first contactless sensor comprises a time-of-flight (ToF) infrared sensor and the second contactless sensor comprises a magnetometer.
14 . The non-transitory computer-readable medium of claim 12 , comprising instructions to match the headphones detected in the first and second sensor data to a previously-paired audio output device.
15 . The non-transitory computer-readable medium of claim 12 , comprising instructions to:
generate aggregated data based on the first and second sensor data; and
process the aggregated data using a machine learning model to generate output indicative of the detected headphones.
16 . The device of claim 1 , wherein the analysis of the ToF sensor data outputs a binary or probabilistic output indicative of the presence of an audio output device.
17 . The device of claim 1 , wherein a portion of the ToF sensor data is cropped to reduce data complexity, consumption of computing resources, or noise, or a combination thereof.
18 . The device of claim 1 , wherein the device comprises a laptop computer, tablet computer, desktop computer, or mobile phone.
19 . The non-transitory computer-readable medium of claim 7 , wherein the machine learning model comprises a convolutional neural network (CNN) comprising a first CNN, the output comprises first output, and the medium comprises instructions to process the depth sensor data based on a second CNN to generate second output indicative of a location of the individual's head.
20 . The non-transitory computer-readable medium of claim 12 , wherein headphones are detected by one of the first or second sensor and not by the other, and the individual is prompted for input soliciting permission to output the audio signal to the wearable audio output device.