# US 2019/0244624 A1 · Audio fingerprinting

> Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt

- Number: US 2019/0244624 A1
- Document ID: US20190244624A1
- Status: Published application
- Kind: A1
- Application number: 16/270,113
- Filed: 2019-02-07
- Published: 2019-08-08
- Earliest claimed: 2013-12-16
- Assignees: Gracenote, Inc.
- Applicants: Gracenote, Inc.
- Inventors: Jinyu Han; Robert Coover
- Web page: https://patentgrep.com/patent/US20190244624A1

## CPC

- G10L 19/018: Audio watermarking, i.e. embedding inaudible data in the audio signal
- G10L: SPEECH ANALYSIS TECHNIQUES OR SPEECH SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING TECHNIQUES; SPEECH OR AUDIO CODING OR DECODING

## Abstract

A machine may be configured to generate one or more audio fingerprints of one or more segments of audio data. The machine may access audio data to be fingerprinted and divide the audio data into segments. For any given segment, the machine may generate a spectral representation from the segment; generate a vector from the spectral representation; generate an ordered set of permutations of the vector; generate an ordered set of numbers from the permutations of the vector; and generate a fingerprint of the segment of the audio data, which may be considered a sub-fingerprint of the audio data. In addition, the machine or a separate device may be configured to determine a likelihood that candidate audio data matches reference audio data.

## Claims

### Claim 1

(canceled)

### Claim 2

An apparatus comprising:
a vector generator to:
determine, first and second groups of frequencies in a plurality of frequencies from spectral data derived from audio data, the first group including frequencies different from frequencies in the second group of frequencies, identify a first subgroup of frequencies in the first group of frequencies based on energy values of the first group,
identify a second subgroup of frequencies in the second group of frequencies based on energy values of the second group, and
generate a vector that assigns a first value to frequencies in the first subgroup and assigns a second value to frequencies in the second subgroup;
a scrambler to generate permutations of the vector, the permutations differently arranging instances of the first and second values;
a coder to generate a sequence that indicates an instance of the first value or of the second value within a corresponding permutation of the permutations; and
a fingerprint generator to generate a fingerprint of the audio data based on the sequence, wherein the generation and decoding of the fingerprint is to conserve computing resources.

### Claim 3 (depends on 2)

The apparatus as defined in claim 2, wherein the first and second values are equal to a shared common value, and wherein the vector assigns the shared common value to frequencies in the first and second subgroups of frequencies.

### Claim 4 (depends on 2)

The apparatus as defined in claim 2, wherein frequencies in the spectral data include a different ordinal position within the spectral data; and, wherein the vector module is to define weighting ones of the respective energy values based on an ordinal position of its corresponding frequency in the spectral data.

### Claim 5 (depends on 4)

The apparatus as defined in claim 4, wherein the weighting of ones of the respective energy values includes multiplying ones of the respective energy values by a corresponding weight factor that indicates the ordinal position of its corresponding frequency in the spectral data.

### Claim 6 (depends on 2)

The apparatus as defined in claim 2, wherein the vector generator is to identify the first subgroup of frequencies based on ranked energy values of the first group of frequencies, and wherein the vector module identifies the second subgroup of frequencies based on ranked energy values of the second group of frequencies.

### Claim 7 (depends on 2)

The apparatus as defined in claim 2, wherein the coder is to generate the sequence by generating an ordered plurality of permutations that arrange the vector different from one another.

### Claim 8 (depends on 2)

The apparatus as defined in claim 2, wherein the coder is to generate the sequences by generating numbers based on calculating a remainder from a modulo operation performed on a numerical representation of a lowest relative position occupied by any instance of the first or second values in the corresponding permutation.

### Claim 9 (depends on 2)

The apparatus as defined in claim 2, wherein the fingerprint module is to generate the fingerprint by storing the sequence with a timestamp that indicates the audio data being fingerprinted.

### Claim 10

A method comprising:
determining, by executing an instruction with at least one processor, first and second groups of frequencies in a plurality of frequencies from spectral data derived from audio data, the first group including frequencies different from frequencies in the second group of frequencies;
identifying, by executing an instruction with the at least one processor, a first subgroup of frequencies in the first group of frequencies based on energy values of the first group;
identifying, by executing an instruction with the at least one processor, a second subgroup of frequencies in the second group of frequencies based on energy values of the second group;
creating, by executing an instruction with the at least one processor, a vector that assigns a first value to frequencies in the first subgroup and assigns a second value to frequencies in the second subgroup;
generating, by executing an instruction with the at least one processor, permutations of the vector, the permutations differently arranging instances of the first and second values;
generating, by executing an instruction with the at least one processor, a sequence that indicates an instance of the first value or of the second value within a corresponding permutation of the permutations; and
generating, by executing an instruction with the at least one processor, a fingerprint of the audio data based on the sequence, wherein the generation and decoding of the fingerprint is to conserve computing resources.

### Claim 11 (depends on 10)

The method as defined in claim 10, wherein the identifying of the first subgroup of frequencies is based on ranked energy values for the first group of frequencies, and wherein the identifying of the second subgroup of frequencies is based on ranked energy values for the second group of frequencies.

### Claim 12 (depends on 10)

The method as defined in claim 10, wherein the generating the sequence includes generating numbers by calculating a remainder from a modulo operation performed on a numerical representation of a lowest relative position occupied by any instance of the first or second values in the corresponding permutation.

### Claim 13 (depends on 10)

The method as defined in claim 10, wherein the generating the fingerprint of the audio data includes storing the sequence with a timestamp that indicates the audio data being fingerprinted.

### Claim 14 (depends on 10)

The method as defined in claim 10, wherein the generating of the fingerprint of the audio data includes storing ones of multiple portions of the sequence in a different corresponding hash table among multiple hash tables that correspond to a timestamp that indicates the audio data being fingerprinted.

### Claim 15

A method comprising:
generating, by executing an instruction with at least one processor, a candidate fingerprint of a candidate audio file by:
determining a first group of frequencies and a second group of frequencies in a plurality of frequencies of spectral data of the candidate audio file, the first group different from the second group,
in the first group of frequencies, identifying a first subgroup of frequencies based on energy values of the first group of frequencies,
in the second group of frequencies, identifying a second subgroup of frequencies based on energy values of the second group,
creating a vector that assigns a first value to frequencies in the first subgroup and assigns a second value to frequencies in the second subgroup,
generating permutations of the vector, the permutations differently arranging instances of the first and second values,
generating a sequence that indicates an instance of the first value or of the second value within a corresponding permutation of the permutations, and
generating the fingerprint based on the sequence; and
comparing, by executing an instruction with the at least one processor, the candidate fingerprint to a reference audio data segment fingerprint.

### Claim 16

An apparatus comprising:
means for identifying first and second groups of frequencies of spectral data derived from audio data;
means for identifying first and second subgroups of first and second groups, respectively;
means for generating a vector that assigns a first value to frequencies of the first group and a second value to frequencies in the second subgroup;
means for generating permutations of the vector; and
means for generating a sequence that indicates an instance of the first value or the second value within a corresponding permutation of the permutations to generate a fingerprint of the audio data based on the sequence.

### Claim 17 (depends on 16)

The apparatus as defined in claim 16, further including means for comparing the fingerprint to candidate audio.

### Claim 18 (depends on 17)

The apparatus as defined in claim 17, further including means for generating a fingerprint of the candidate audio.

### Claim 19 (depends on 16)

The apparatus as defined in claim 16, further including means for weighting energy values of the spectral data.

### Claim 20 (depends on 16)

The apparatus as defined in claim 16, wherein the means for generating the sequence includes means for ordering the permutations.

### Claim 21

A non-transitory machine readable medium comprising instructions, which when executed, cause a processor to at least:
determine first and second groups of frequencies in a plurality of frequencies from spectral data derived from audio data, the first group including frequencies different from frequencies in the second group of frequencies;
identify a first subgroup of frequencies in the first group of frequencies based on energy values of the first group;
identify a second subgroup of frequencies in the second group of frequencies based on energy values of the second group;
create a vector that assigns a first value to frequencies in the first subgroup and assigns a second value to frequencies in the second subgroup;
generate permutations of the vector, the permutations differently arranging instances of the first and second values;
generate a sequence that indicates an instance of the first value or of the second value within a corresponding permutation of the permutations; and
generate a fingerprint of the audio data based on the sequence.

### Claim 22 (depends on 21)

The non-transitory machine readable medium as defined in claim 21, wherein the first subgroup of frequencies is identified based on ranked energy values for the first group of frequencies, and wherein the identifying of the second subgroup of frequencies is identified based on ranked energy values for the second group of frequencies.

### Claim 23 (depends on 21)

The non-transitory machine readable medium as defined in claim 21, wherein the sequence is generated by generating numbers based on calculating a remainder from a modulo operation performed on a numerical representation of a lowest relative position occupied by any instance of the first or second values in the corresponding permutation.

### Claim 24 (depends on 21)

The non-transitory machine readable medium as defined in claim 21, wherein the fingerprint is generated by storing ones of multiple portions of the sequence in a different corresponding hash table among multiple hash tables that correspond to a timestamp that indicates the audio data being fingerprinted
