AUDIO SYSTEM INCLUDING FOR NEAR FIELD AND FAR FIELD ENHANCEMENT THAT USES A CONTACT TRANSDUCER
An audio system for near and far field signal enhancement that uses a contact transducer. The audio system includes a microphone array, the contact transducer, and a controller. The microphone array detects sounds from a local area. The sounds from the local area include a voice of a user of the audio system. The contact transducer may be in contact with the head of the user. The contact transducer detects tissue based vibrations that are generated by the user and pass through tissue of the user prior to being detected by the contact transducer. The controller identifies the voice of the user in the detected sounds using the detected tissue based vibrations and a model, and updates a sound filter based on the identified voice of the user. The audio content is modified using the updated sound filter, and is presented by at least one audio system.
1 . An audio system comprising:
a microphone array configured to detect sounds from a local area, the sounds from the local area including a voice of a user of the audio system;
a contact transducer configured to detect tissue based vibrations on a portion of a head of the user, the vibrations generated by the user and pass through tissue of the user prior to being detected by the contact transducer;
a controller configured to:
identify the voice of the user in the detected sounds using the detected tissue based vibrations and a model; and
update a sound filter based on the identified voice of the user,
wherein audio content is modified using the updated sound filter, and the modified audio content is presented by at least one audio system.
2 . The audio system of claim 1 , wherein the audio system is integrated into a headset.
3 . The audio system of claim 2 , wherein the contact transducer is configured to sense vibrations of a portion of a nose of the user.
4 . The audio system of claim 1 , wherein the tissue based vibrations are caused by the voice of the user, and the updated sound filter enhances the voice of the user, and the controller is further configured to:
modify the audio content with the updated filter, wherein the modified audio content enhances the voice of the user; and
provide the modified audio content to a second audio system, wherein the second audio system presents the modified audio content.
5 . The audio system of claim 1 , wherein the tissue based vibrations are caused by voice of the user, and the updated sound filter enhances the voice of the user, and the controller is further configured to:
modify the audio content with the updated filter, wherein the modified audio content enhances the voice of the user;
determine that the modified audio content includes a command; and
perform an action in accordance with the command.
6 . The audio system of claim 1 , wherein the tissue based vibrations are caused by voice of the user, and the controller is further configured to:
train an adaptive beamformer using the tissue based vibrations and the sounds from the local area.
7 . The audio system of claim 1 , wherein the tissue based vibrations are caused by voice of the user, and the controller is further configured to:
determine spectral and spatial correlations between the voice of the user and the sounds from the local area using the tissue based vibrations; and
train the learning model using the determined correlations to distinguish between the voice of the user and other sounds from the local area.
8 . The audio system of claim 1 , wherein the tissue based vibrations are caused by voice of the user, and the controller is further configured to:
determine one or more functions describing the voice within the local area using the tissue based vibrations, wherein the functions are selected from a group comprising: a temporal response of the voice within the local area, a spectral response of the voice within the local area, and a spatial response of the voice within the local area;
train the model using the determined one or more functions to distinguish between the voice of the user and other sounds from the local area.
9 . A method comprising:
detecting, via a microphone array of an audio system, sounds from a local area, the sounds from the local area including a voice of a user of the audio system;
detecting, via a contact transducer, tissue based vibrations on a portion of a head of the user, the vibrations are generated by the user and pass through tissue of the user prior to being detected by the contact transducer;
identifying the voice of the user in the detected sounds using the detected tissue based vibrations and a model; and
updating a sound filter based on the identified voice of the user,
wherein audio content is modified using the updated sound filter, and the modified audio content is presented by at least one audio system.
10 . The method of claim 9 , wherein the tissue based vibrations are caused by the voice of the user, and the updated sound filter enhances the voice of the user, and the method further comprises:
modifying the audio content with the updated filter, wherein the modified audio content enhances the voice of the user; and
providing the modified audio content to a second audio system, wherein the second audio system presents the modified audio content.
11 . The method of claim 9 , wherein the tissue based vibrations are caused by voice of the user, and the updated sound filter enhances the voice of the user, and the method further comprises:
modifying the audio content with the updated filter, wherein the modified audio content enhances the voice of the user;
determining that the modified audio content includes a command; and
performing an action in accordance with the command.
12 . The method of claim 9 , wherein the tissue based vibrations are caused by voice of the user, and the method further comprises:
training an adaptive beamformer using the tissue based vibrations and the sounds from the local area.
13 . The method of claim 9 , wherein the tissue based vibrations are caused by voice of the user, and the method further comprising:
determining spectral and spatial correlations between the voice of the user and the sounds from the local area using the tissue based vibrations; and
training the learning model using the determined correlations to distinguish between the voice of the user and other sounds from the local area.
14 . The method of claim 1 , wherein the tissue based vibrations are caused by voice of the user, and the method further comprising:
determining one or more functions describing the voice within the local area using the tissue based vibrations, wherein the functions are selected from a group comprising: a temporal response of the voice within the local area, a spectral response of the voice within the local area, and a spatial response of the voice within the local area;
training the model using the determined one or more functions to distinguish between the voice of the user and other sounds from the local area.
15 . The method of claim 9 , wherein the audio system is integrated into a headset.
16 . The method of claim 15 , wherein the contact transducer is configured to be in contact with a portion of a nose of the user.
17 . A non-transitory computer readable medium configured to store program code instructions, when executed by a processor of an audio system, cause the audio system to perform steps comprising:
detecting, via a microphone array of an audio system, sounds from a local area, the sounds from the local area including a voice of a user of the audio system;
detecting, via a contact transducer, tissue based vibrations on a portion of a head of the user, the vibrations are generated by the user and pass through tissue of the user prior to being detected by the contact transducer;
identifying the voice of the user in the detected sounds using the detected tissue based vibrations and a model; and
updating a sound filter based on the identified voice of the user,
wherein audio content is modified using the updated sound filter, and the modified audio content is presented by at least one audio system.
18 . The computer readable medium of claim 17 , wherein the tissue based vibrations are caused by voice of the user, and the program code instructions, when executed by the processor, further cause the processer to perform steps comprising:
training an adaptive beamformer using the tissue based vibrations and the sounds from the local area.
19 . The computer readable medium of claim 17 , wherein the tissue based vibrations are caused by voice of the user, and the program code instructions, when executed by the processor, further cause the processer to perform steps comprising:
determining spectral and spatial correlations between the voice of the user and the sounds from the local area using the tissue based vibrations; and
training the learning model using the determined correlations to distinguish between the voice of the user and other sounds from the local area.
20 . The computer readable medium of claim 17 , wherein the tissue based vibrations are caused by voice of the user, and the and the program code instructions, when executed by the processor, further cause the processer to perform steps comprising:
determining one or more functions describing the voice within the local area using the tissue based vibrations, wherein the functions are selected from a group comprising: a temporal response of the voice within the local area, a spectral response of the voice within the local area, and a spatial response of the voice within the local area;
training the model using the determined one or more functions to distinguish between the voice of the user and other sounds from the local area.