Sound recognition

Patent metadata

Field	Value
Publication number	US-2017195810-A1
Application number	US-201514986243-A
Country	US
Kind code	A1
Filing date	Dec 31, 2015
Priority date	Dec 31, 2015
Publication date	Jul 6, 2017
Grant date	—

Abstract

Methods, systems, and apparatus for monitoring a sound are described. An audio signal is obtained and the audio signal is analyzed to generate an audio signature. An object type is identified based on the audio signature and an action corresponding to the object type is identified.

First claim

What is claimed is:

1 . An apparatus for monitoring and identifying a source of a sound, the apparatus comprising:
an audio analysis module, implemented using a hardware device, for obtaining an audio signal and analyzing the audio signal to generate an audio signature; a signature analysis module for identifying an object type based on the audio signature; and an action module for identifying an action corresponding to the object type.

2 . The apparatus of claim 1 , wherein the signature analysis module is further configured to identify the object based on the audio signature.

3 . The apparatus of claim 1 , wherein the signature analysis module is further configured to identify a characteristic of the object based on the audio signature.

4 . The apparatus of claim 1 , further comprising a training module for storing the audio signature in a signature library and generating an index according to the object type.

5 . The apparatus of claim 1 , further comprising a training module for storing the audio signature in a signature library and generating an index according to the identity of the object or a characteristic of the object.

6 . The apparatus of claim 1 , wherein the object type is identified by comparing the audio signature to one or more signatures in a library of signatures.

7 . The apparatus of claim 1 , wherein the determined action is based on one or more rules, each rule defining an action to recommend or perform based on the audio signature.

8 . The apparatus of claim 1 , further comprising an audio pre-processing module to isolate one or more sounds in the audio signal.

9 . The apparatus of claim 1 , wherein a sound is monitored periodically, at a prescheduled time, in response to a reception of a measurement request, and in response to a substantial change in the audio signal.

10 . The apparatus of claim 1 , wherein the analyzing operation analyzes the sound emitted by the object to determine a behavior of a user.

11 . The apparatus of claim 1 , wherein the analyzing operation is based on characteristics of an audio waveform, including a frequency range, a tempo, acoustic resonances, or a power in each of a plurality of frequency ranges.

12 . A method for monitoring and identifying a source of a sound, the method comprising:
obtaining an audio signal and analyzing the audio signal to generate an audio signature; identifying an object type based on the audio signature; and identifying an action corresponding to the object type.

13 . The method of claim 12 , further comprising identifying the object based on the audio signature.

14 . The method of claim 12 , further comprising identifying a characteristic of the object based on the audio signature.

15 . The method of claim 12 , further comprising storing the audio signature in a signature library and generating an index according to the object type.

16 . The method of claim 12 , further comprising storing the audio signature in a signature library and generating an index according to the identity of the object or a characteristic of the object.

17 . The method of claim 12 , wherein the object type is identified by comparing the audio signature to one or more signatures in a library of signatures.

18 . The method of claim 12 , wherein the determined action is based on one or more rules, each rule defining an action to recommend or perform based on the audio signature.

19 . The method of claim 12 , wherein the analyzing operation is based on characteristics of an audio waveform, including a frequency range, a tempo, acoustic resonances, or a power in each of a plurality of frequency ranges.

20 . A non-transitory computer-readable medium embodying instructions that, when executed by a processor of a machine, cause the machine to perform operations comprising:
obtaining an audio signal and analyzing the audio signal to generate an audio signature; identifying an object type based on the audio signature; and identifying an action corresponding to the object type.

Assignees

• Ebay Inc

Inventors

• Gonzales Jr Sergio Pinzon

Classifications

• G10L25/18

  the extracted parameters being spectral information of each sub-band · CPC title

• G07C5/0808 Primary

  Diagnosing performance data (testing of vehicles G01M17/00; testing of electrical installation on vehicles G01R31/005) · CPC title

• G07C5/006

  Indicating maintenance · CPC title

• G06F16/638

  Presentation of query results · CPC title

• G10L25/51

  for comparison or discrimination · CPC title