User voice control system
Embodiments include techniques and objects related to a wearable audio device that includes a microphone to detect a plurality of sounds in an environment in which the wearable audio device is located. The wearable audio device further includes a non-acoustic sensor to detect that a user of the wearable audio device is speaking. The wearable audio device further includes one or more processors communicatively to alter, based on an identification by the non-acoustic sensor that the user of the wearable audio device is speaking, one or more of the plurality of sounds to generate a sound output. Other embodiments may be described or claimed.
1. A method comprising:
identifying, by a wearable audio device, a plurality of sounds in an environment in which the wearable audio device is located, wherein the plurality of sounds are captured by a microphone of the wearable audio device;
identifying, by a model of a neural network related to the wearable audio device, that the plurality of sounds includes a voice of a user that is wearing the wearable audio device; and
altering or facilitating alteration of, by the model in response to identification that the plurality of sounds includes the voice of the user, playback of the plurality of sounds to the user, wherein altering playback of the plurality of sounds to the user includes decreasing a volume of playback of the voice of the user.
2. The method of claim 1 , wherein alteration of playback of the plurality of sounds includes decreasing volume of one or more sounds of the plurality of sounds.
3. The method of claim 1 , wherein alteration of playback of the plurality of sounds includes altering sounds in a first pre-identified frequency band and not altering sounds in a second pre-identified frequency band.
4. The method of claim 1 , wherein alteration of playback of the plurality of sounds includes altering volume of all sounds of the plurality of sounds.
5. The method of claim 1 , wherein
identification that the plurality of sounds includes the voice of the user is based on detection that the user is speaking by a sensor of the wearable audio device.
6. The method of claim 5 , wherein the sensor is an accelerometer, a piezoelectric sensor, or a vibration sensor.
7. One or more non-transitory computer-readable media comprising instructions that, upon execution of the instructions by one or more processors of a wearable audio device, are to cause the wearable audio device to:
identify, based on a non-acoustic sensor that is in physical contact with a user that is wearing the wearable audio device, that the user is speaking;
alter, based on the identification that the user is speaking, a muted audio recording recorded by a microphone of the wearable audio device, wherein the audio recording is related to an environment in which the wearable audio device is located, and wherein the alteration of the audio recording includes:
un-muting sounds in a first frequency band of the audio recording that is related to a voice of the user; and
not un-muting sounds in a second frequency band of the audio recording that is unrelated to the voice of the user; and
output the audio recording.
8. The one or more non-transitory computer-readable media of claim 7 , wherein alteration of the audio recording includes un-muting the audio recording.
9. The one or more non-transitory computer-readable media of claim 7 , wherein the non-acoustic sensor is an accelerometer.
10. The one or more non-transitory computer-readable media of claim 7 , wherein the non-acoustic sensor is a vibration sensor.
11. The one or more non-transitory computer-readable media of claim 7 , wherein the non-acoustic sensor is a piezoelectric sensor.
12. The one or more non-transitory computer-readable media of claim 7 , wherein the alteration of the audio recording is performed by a neural network.
13. A system comprising:
a wearable audio device that includes logic configured to identify a plurality of sounds in an environment in which the wearable audio device is located, wherein the plurality of sounds are captured by a microphone of the wearable audio device; and
logic to implement a neural network, wherein the network is configured to generate a model that is operable to:
identify that the plurality of sounds includes a voice of a user that is wearing the wearable audio device; and
alter or facilitate alteration of, in response to identification that the plurality of sounds includes the voice of the user, playback of the plurality of sounds to the user, wherein altering playback of the plurality of sounds to the user includes decreasing a volume of playback of the voice of the user.
14. The system of claim 13 , wherein alteration of playback of the plurality of sounds includes decreasing volume of one or more sounds of the plurality of sounds.
15. The system of claim 13 , wherein alteration of playback of the plurality of sounds includes altering sounds in a first pre-identified frequency band and not altering sounds in a second pre-identified frequency band.
16. The system of claim 13 , wherein alteration of playback of the plurality of sounds includes altering volume of all sounds of the plurality of sounds.
17. The system of claim 13 , wherein identification that the plurality of sounds includes the voice of the user is based on detection that the user is speaking by a sensor of the wearable audio device.
18. The system of claim 17 , wherein the sensor is an accelerometer, a piezoelectric sensor, or a vibration sensor.
19. The system of claim 13 , wherein the logic to implement the neural network is an element of the wearable audio device.