CROWDSOURCED DATABASE FOR SOUND IDENTIFICATION

• US 20170194021A1
• Filed: 12/31/2015
• Published: 07/06/2017
• Est. Priority Date: 12/31/2015
• Status: Active Grant

• Current Assignee: Harman International Industries Incorporated (Samsung Electronics Co. Ltd.)
• Original Assignee: Harman International Industries Incorporated (Samsung Electronics Co. Ltd.)
• Inventors: Butts, Donald Joseph, Stacey, Brandon
• Granted Patent: US 9,830,931 B2
• Time in Patent Office: Days
• CPC Class Codes: G06F 16/61 Indexing; Data structures therefor; Storage structures
• CPC Class Codes: G10L 25/51 for comparison or discrimination
• CPC Class Codes: G10L 25/72 for transmitting results of analysis
• CPC Class Codes: H04B 15/00 Suppression or limitation of noise or interference by means associated with receiver H04B1/10
• CPC Class Codes: H04R 1/1008 Earpieces of the supra-aural or circum-aural type
• CPC Class Codes: H04R 1/1041 Mechanical or electronic switches, or control elements switches in general H01H
• CPC Class Codes: H04R 2225/41 Detection or adaptation of hearing aid parameters or programs to listening situation, e.g. pub, forest
• CPC Class Codes: H04R 2430/01 Aspects of volume control, not necessarily automatic, in sound systems
• CPC Class Codes: H04R 25/552 Binaural
• CPC Class Codes: H04R 25/554 using a wireless connection, e.g. between microphone and amplifier or using Tcoils
• CPC Class Codes: H04R 3/00 Circuits for transducers , loudspeakers or microphones
• CPC Class Codes: H04R 5/033 Headphones for stereophonic communication details thereof, e.g. relating to batteries, cables or control elements H04R1/10
• CPC Class Codes: H04R 5/04 Circuit arrangements, e.g. for selective connection of amplifier inputs/outputs to loudspeakers, for loudspeaker detection, or for adaptation of settings to personal preferences or hearing impairments combinations of amplifiers H03F3/68; stereophonic systems H04S

Abstract

One embodiment of the present invention sets forth a technique for determining a set of sound parameters associated with a sound type. The technique includes receiving, via a network and from each a first plurality of remote computing devices, an audio recording of a first sound type and a descriptor associated with the first sound type. The technique further includes processing the audio recordings via a processor to determine a first set of sound parameters associated with the first sound type. The technique further includes receiving a request associated with the descriptor from at least one remote computing device and, in response, transmitting the first set of sound parameters associated with the first sound type to the at least one remote computing device.

Claims

1. A non-transitory computer-readable storage medium including instructions that, when executed by a processor, configure the processor to determine a set of sound parameters associated with a sound type, by performing the steps of:

receiving, via a network, a different audio recording of a first sound type from each remote computing device included in a first plurality of remote computing devices;

processing the audio recordings via a processor to determine a first set of sound parameters associated with the first sound type; and

transmitting the first set of sound parameters associated with the first sound type to at least one remote computing device.

2. The non-transitory computer-readable storage medium of claim 1, wherein the instructions further configure the processor to receive a request for the first set of sound parameters from the at least one remote computing device, wherein the first set of sound parameters configures the at least one remote computing device to identify the first sound type within ambient sound.

3. The non-transitory computer-readable storage medium of claim 1, wherein processing the audio recordings comprises applying a machine learning algorithm to the audio recordings of the first sound type to separate the first sound type from noise included in each of the audio recordings.

4. The non-transitory computer-readable storage medium of claim 3, wherein the machine learning algorithm comprises independent component analysis (ICA).

5. The non-transitory computer-readable storage medium of claim 1, wherein processing the audio recordings of the first sound type comprises minimizing a cost function to train a source separation algorithm.

6. The non-transitory computer-readable storage medium of claim 1, wherein the instructions further configure the processor to:

receive, via the network, a different audio recording of a second sound type from each remote computing device included in a second plurality of remote computing devices;

process the audio recordings of the second sound type via the processor to determine a second set of sound parameters associated with the second sound type; and

transmit the second set of sound parameters associated with the second sound type to at least one remote computing device.

7. The non-transitory computer-readable storage medium of claim 6, wherein the instructions further configure the processor to:

receive, from each remote computing device included in the first plurality of remote computing devices, a first descriptor associated with the first sound type;

receive, from each remote computing device included in the second plurality of remote computing devices, a second descriptor associated with the second sound type;

sort, based on the first descriptors and the second descriptors, the audio recordings received from the first plurality of remote computing devices and the second plurality of remote computing devices into a first group of audio recordings associated with the first sound type and a second group of audio recordings associated with the second sound type;

associate the first set of sound parameters with the first descriptor; and

associate the second set of sound parameters with the second descriptor.

8. The non-transitory computer-readable storage medium of claim 7, wherein each of the first descriptor and the second descriptor comprises at least one of a product brand and a product model.

9. The non-transitory computer-readable storage medium of claim 1, wherein the remote computing devices included in the first plurality of remote computing devices comprise at least one of a mobile computer, a wireless headphone, and a media player.

10. A method for determining a set of sound parameters associated with a sound type, the method comprising:

receiving, via a network and from each remote computing device included in a first plurality of remote computing devices, an audio recording of a first sound type and a descriptor associated with the first sound type;

processing the audio recordings via a processor to determine a first set of sound parameters associated with the first sound type;

receiving a request associated with the descriptor from at least one remote computing device; and

in response, transmitting the first set of sound parameters associated with the first sound type to the at least one remote computing device.

11. The method of claim 10, wherein the first set of sound parameters configures the at least one remote computing device to identify the first sound type within ambient sound.

12. The method of claim 10, further comprising:

receiving, via the network and from each remote computing device included in a second plurality of remote computing devices, an additional audio recording of a first sound type and the descriptor associated with the first sound type; and

processing the additional audio recordings via the processor to determine an updated set of sound parameters associated with the first sound type.

13. The method of claim 12, further comprising transmitting the updated set of sound parameters associated with the first sound type to the at least one remote computing device.

14. The method of claim 10, wherein processing the audio recordings comprises training a source separation algorithm to separate the first sound type from noise included in each of the audio recordings.

15. The method of claim 10, wherein the first plurality of remote computing devices comprise at least one of a mobile computer, a pair of headphones, and a media player.

17. The method of claim 10, further comprising, after receiving the audio recordings of the first sound type, storing the audio recordings in a cloud storage device, wherein the first set of sound parameters are transmitted from the cloud storage device to the remote computing device in response to receiving the request associated with the descriptor.

18. The method of claim 10, wherein receiving the audio recordings of the first sound type comprises receiving at least one-hundred different audio recordings.

16. The method of claim 16, wherein the first descriptor comprises at least one of a vehicle model, an appliance model, a name of a person, and an animal.

19. A system for determining a set of sound parameters associated with a sound type, the system comprising:

a memory storing a crowdsource application; and

a processor that is coupled to the memory and, when executing the crowdsource application, is configured to;

receive, via a network, a different audio recording of a first sound type from each remote computing device included in a first plurality of remote computing devices;

process the audio recordings to separate the first sound type from noise included in each of the audio recordings;

determine a first set of sound parameters associated with the first sound type based on the audio recordings that have been processed; and

transmit the first set of sound parameters associated with the first sound type to at least one remote computing device.

20. The system of claim 19, wherein the crowdsource application further configures the processor to store the audio recordings of the first sound type and the first set of sound parameters in a cloud storage device, and wherein the first set of sound parameters are transmitted from the cloud storage device to the at least one remote computing device.