Methods and apparatus for determining a mood profile associated with media data
View Patent ↗In an embodiment, a method is provided for determining a mood profile of media data. In this method, mood is determined across multiple elements of mood for the media data to create a mood profile associated with the media data. In some embodiments, the mood profile is then used to determine congruencies between one or more pieces of media data.
1. A method of identifying a congruency of media data, the method comprising:
accessing a first mood profile digest comprising a first vector including a first plurality of first mood categories, each having an associated value;
accessing a second mood profile digest comprising a second vector including a second plurality of second mood categories, each having an associated value;
determining that at least one first mood category is similar to at least one second mood category based on a correlates matrix that quantifies a similarity relationship between the first mood category and the second mood category,
the determining including calculation of a normalized comparison vector of the first mood profile digest in accordance with
A
comp
[
n
]
=
∑
m
=
1
M
A
score
[
m
]
×
C
(
A
category
[
m
]
,
A
category
[
n
]
)
,
wherein A comp [n] is an n th element of the normalized comparison vector A comp , A score [m] is a score of an m th element of the first mood profile digest A, A category [m] is a first mood category of the m th element of the first mood profile digest A, and C(x,y) is a comparison value between the first and second mood categories x and y as indicated in the correlates matrix; and
dependent upon the similarity between the first mood category and the second mood category,
comparing the value associated with the first mood category with the value associated with the second mood category,
the comparing of the values being performed by a processor of a machine; and
generating a similarity score representing a degree of congruency between the first mood profile digest and the second mood profile digest based on the comparing of the associated values.
2. The method of claim 1 , wherein
the first mood profile digest is a portion of a first mood profile; and
the second mood profile digest is a portion of a second mood profile.
3. The method of claim 1 , wherein
the first mood profile digest is a summarization of a first mood profile; and
the second mood profile digest a summarization of a second mood profile.
4. The method of claim 1 further comprising:
comparing a further value associated with a further first mood category of the first plurality of first mood categories with a further value associated with a further second mood category of the second plurality of second mood categories; and wherein
the generating of the similarity score is based on the comparing of the further values.
5. The method of claim 1 , wherein
the correlates matrix contains a comparative value indicating an extent of similarity between the first mood category and the second mood category; and wherein
the comparing of the value associated with the first mood category with the value associated with the second mood category is based on the comparative value.
6. The method of claim 5 , further comprising:
generating a first set of normalized values based on the comparative value and based on values associated with the first plurality of first mood categories;
generating a second set of normalized values based on the comparative value and based on values associated with the second plurality of second mood categories; and wherein
the generating of the similarity score is based on the first set of normalized values and based on the second set of normalized values.
7. The method of claim 6 , further comprising
determining that the first mood category and the second mood category are different mood categories; and wherein
the generating of at least one of the first set of normalized values or the second set of normalized values is based on the determining that the first mood category in the second mood category are not the same.
8. The method of claim 6 , wherein
the first mood category is predominant among the first plurality of first mood categories, based on the value associated with the first mood category being largest within the first vector;
the second mood category is predominant among the second plurality of second mood categories, based on the value associated with the second mood category being largest within the second vector; and the method further comprises
identifying the comparative value within the correlates matrix based on the first and second mood categories being predominant.
9. The method of claim 1 , wherein:
the first mood profile digest is associated with a first recording;
the second mood profile digest is associated with a second recording; and the method further comprises
retrieving the second recording from a database based on the similarity score exceeding a threshold similarity.
10. The method of claim 1 , further comprising:
communicating an audio identifier to a server,
the audio identifier identifying an audio recording accessible by a media player;
receiving the first mood profile digest from the server,
the first mood profile digest identifying at least one mood associated with the audio recording; and
selecting the second mood profile digest from a set of mood profile digests received from the server.
11. A non-transitory machine-readable storage medium comprising instructions that, when executed by one or more processors of a machine, cause the machine to perform operations comprising:
accessing a first mood profile digest comprising a first vector including a first plurality of first mood categories, each having an associated value;
accessing a second mood profile digest comprising a second vector including a second plurality of second mood categories, each having an associated value;
determining that at least one first mood category is similar to at least one second mood category based on a correlates matrix that quantifies a similarity relationship between the first mood category and the second mood category,
the determining including calculation of a normalized comparison vector of the first mood profile digest in accordance with
A
comp
[
n
]
=
∑
m
=
1
M
A
score
[
m
]
×
C
(
A
category
[
m
]
,
A
category
[
n
]
)
,
wherein A comp [n] is an n th element of the normalized comparison vector A comp , A score [m] is a score of an m th element of the first mood profile digest A, A category [m] is a first mood category of the m th element of the first mood profile digest A, and C(x,y) is a comparison value between the first and second mood categories x and y as indicated in the correlates matrix; and
dependent upon the similarity between the first mood category and the second mood category,
comparing the value associated with the first mood category with the value associated with the second mood category, the comparing of the values being performed by the one or more processors of the machine; and
generating a similarity score representing a degree of congruency between the first mood profile digest and the second mood profile digest based on the comparing of the associated values.
12. The non-transitory machine-readable storage medium of claim 11 , wherein:
the correlates matrix contains a comparative value indicating an extent of similarity between the first mood category and the second mood category; and wherein
the comparing of the value associated with the first mood category with the value associated with the second mood category is based on the comparative value.
13. The non-transitory machine-readable storage medium of claim 12 , wherein the operations further comprise:
generating a first set of normalized values based on the comparative value and based on values associated with the first plurality of first mood categories;
generating a second set of normalized values based on the comparative value and based on values associated with the second plurality of second mood categories; and wherein
the generating of the similarity score is based on the first set of normalized values and based on the second set of normalized values.
14. The non-transitory machine-readable storage medium of claim 13 , wherein the operations further comprise:
determining that the first mood category and the second mood category are different mood categories; and wherein
the generating of at least one of the first set of normalized values or the second set of normalized values is based on the determining that the first mood category in the second mood category are not the same.
15. The non-transitory machine-readable storage medium of claim 13 , wherein:
the first mood category is predominant among the first plurality of first mood categories, based on the value associated with the first mood category being largest within the first vector;
the second mood category is predominant among the second plurality of second mood categories, based on the value associated with the second mood category being largest within the second vector; and the operations further comprise
identifying the comparative value within the correlates matrix based on the first and second mood categories being predominant.
16. Apparatus comprising:
a mood determining module configured to:
access a first mood profile digest comprising a first vector including a first plurality of first mood categories, each having an associated value; and
access a second mood profile digest comprising a second vector including a second plurality of second mood categories, each having an associated value; and
a processor configured by a mood comparing module to:
determine that at least one first mood category is similar to at least one second mood category based on a correlates matrix that quantifies a similarity relationship between the first mood category and the second mood category,
the determining including calculation of a normalized comparison vector of the first mood profile digest in accordance with
A
comp
[
n
]
=
∑
m
=
1
M
A
score
[
m
]
×
C
(
A
category
[
m
]
,
A
category
[
n
]
)
,
wherein A comp [n] is an n th element of the normalized comparison vector A comp, A score [m] is a score of an m th element of the first mood profile digest A, A category [m] is a first mood category of the m th element of the first mood profile digest A, and C(x,y) is a comparison value between the first and second mood categories x and y as indicated in the correlates matrix; and
dependent upon the similarity between the first mood category and the second mood category,
compare the value associated with the first mood category with the value associated with the second mood category; and
generate a similarity score representing a degree of congruency between the first mood profile digest and the second mood profile digest based on the comparing of the associated values.
17. The apparatus of claim 16 , wherein:
the correlates matrix contains a comparative value indicating an extent of similarity between the first mood category and the second mood category; and wherein
the mood comparing module configures the processor to compare the value associated with the first mood category with the value associated with the second mood category based on the comparative value.
18. The apparatus of claim 17 , wherein:
the mood comparing module configures the processor to:
generate a first set of normalized values based on the comparative value and based on values associated with the first plurality of first mood categories; and
generate a second set of normalized values based on the comparative value and based on values associated with the second plurality of second mood categories; and wherein
the generating of the similarity score is based on the first set of normalized values and based on the second set of normalized values.
19. The apparatus of claim 18 , wherein:
the mood comparing module configures the processor to determine that the first mood category and the second mood category are different mood categories; and
the generating of at least one of the first set of normalized values or the second set of normalized values is based on the determining that the first mood category in the second mood category are not the same.
20. The apparatus of claim 18 , wherein:
the first mood category is predominant among the first plurality of first mood categories, based on the value associated with the first mood category being largest within the first vector;
the second mood category is predominant among the second plurality of second mood categories, based on the value associated with the second mood category being largest within the second vector; and
the mood comparing module configures the processor to identify the comparative value within the correlates matrix based on the first and second mood categories being predominant.