Methods and apparatus to identify a mood of media
Methods and apparatus to identify an emotion evoked by media are disclosed. An example apparatus includes a synthesizer to generate a first synthesized sample based on a pre-verbal utterance associated with a first emotion. A feature extractor is to identify a first value of a first feature of the first synthesized sample. The feature extractor to identify a second value of the first feature of first media evoking an unknown emotion. A classification engine is to create a model based on the first feature. The model is to establish a relationship between the first value of the first feature and the first emotion. The classification engine is to identify the first media as evoking the first emotion when the model indicates that the second value corresponds to the first value.
1. An apparatus to identify an emotion evoked by media, the apparatus comprising:
notator circuitry to create a musical representation of a pre-verbal utterance known to evoke a first emotion;
display controller circuitry to cause a display to present an interface to instruct a musician to perform the musical representation of the pre-verbal utterance;
a feature extractor to identify a first value of a first feature of a recording of the performance of the musical representation of the pre-verbal utterance, the feature extractor to identify a second value of the first feature of first media evoking an unknown emotion;
a classification engine to create a model based on the first feature, the model to establish a relationship between the first value of the first feature and the first emotion, the classification engine to identify the unknown emotion as the first emotion when the model indicates that the second value corresponds to the first value; and
a recommendation engine to calculate emotional distances between respective ones of potential media for recommendation and the first emotion, the recommendation engine to select one of the potential media based on the respective emotional distances.
2. The apparatus as described in claim 1 , wherein the feature extractor is to identify a third value of the first feature of second media as evoking the first emotion, and further including a model validator to validate the model by confirming that the model indicates that the first value of the first feature is within a threshold percentage of the third value of the first feature.
3. The apparatus as described in claim 2 , wherein the validator includes a semantic mapper to map a second emotion identified as being evoked by the first media to the first emotion.
4. The apparatus as described in claim 1 , wherein the recommendation engine is to recommend the first media in response to a request for media evoking the first emotion.
5. The apparatus as described in claim 1 , wherein the feature extractor includes at least one of a zero crossing identifier, a rolloff power identifier, a brightness identifier, a roughness identifier, a minor third interval identifier, a major third interval identifier, an irregularity identifier, a chroma identifier, a main pitch identifier, or a key identifier.
6. The apparatus as described in claim 1 , wherein the feature extractor includes at least three of zero crossing identifier, a rolloff power identifier, a brightness identifier, a roughness identifier, a minor third interval identifier, a major third interval identifier, an irregularity identifier, a chroma identifier, a main pitch identifier, or a key identifier.
7. A non-transitory machine readable storage medium comprising instructions which, when executed, cause a machine to at least:
create a musical representation of a pre-verbal utterance known to evoke a first emotion;
cause display of an interface to instruct a musician to perform the musical representation of the pre-verbal utterance;
calculate a first value of a first feature of a recording of the performance of the musical representation of the pre-verbal utterance;
create a model based on the first value of the first feature, the model to establish a relationship between the first value of the first feature and the first emotion;
identify a second value of the first feature of first media evoking an unknown emotion;
identify the unknown emotion as the first emotion when the model indicates that the second value corresponds to the first value;
calculate emotional distances between respective ones of potential media for recommendation and the first emotion; and
recommend one of the potential media based on the respective emotional distances.
8. The non-transitory machine readable storage medium as described in claim 7 , wherein the instructions, when executed, cause the machine to use a synthesized musical instrument to synthesize a first synthesized sample.
9. The non-transitory machine readable storage medium as described in claim 8 , wherein the musical representation of the pre-verbal utterance is a Musical Instrument Digital Interface representation of the pre-verbal utterance.
10. The non-transitory machine readable storage medium as described in claim 9 , wherein the instructions, when executed, cause the machine to:
generate a vocoder representation of the Musical Instrument Digital Interface representation; and
use the vocoder representation to synthesize the first synthesized sample.
11. The non-transitory machine readable storage medium as described in claim 7 , wherein the instructions, when executed, cause the machine to generate a vocoder representation of the recording.
12. The non-transitory machine readable storage medium as described in claim 7 , wherein the instructions, when executed, cause the machine to instruct the musician to emulate the pre-verbal utterance.
13. The non-transitory machine readable storage medium as described in claim 7 , wherein the instructions, when executed, cause the machine to at least:
identify a second value of the first feature of second media evoking the first emotion; and
confirm that the emotion model indicates that the second media corresponds to the first emotion.
14. The non-transitory machine readable storage medium as described in claim 13 , wherein the instructions, when executed, cause the machine to update the model when the second value of the first feature does not correspond to the first value of the first feature.
15. The non-transitory machine readable storage medium as described in claim 7 , wherein the first feature is at least one of a number of zero crossings, a rolloff power, a brightness, a roughness, a presence of a minor third interval, a presence of a major third interval, an irregularity, a chroma, a main pitch, or a key.
16. The non-transitory machine readable storage medium as described in claim 7 , wherein the instructions, when executed, cause the machine to at least:
identify a request for media evoking the first emotion; and
recommend second media as evoking the first emotion.
17. The non-transitory machine readable storage medium as described in claim 7 , wherein the instructions, when executed, cause the machine to identify the first emotion based on an emotion evoked by primary media.
18. A method to identify an emotion evoked by media, the method comprising:
creating, by executing an instruction with a processor, a musical representation of a pre-verbal utterance known to evoke a first emotion;
causing, by executing an instruction with a processor, a display device to present an interface to instruct a musician to perform the musical representation of the pre-verbal utterance;
accessing, by executing an instruction with the processor, a first sample of the musician performing the musical representation of the pre-verbal utterance;
calculating, by executing an instruction with the processor, a first value of a first feature of the first synthesized sample;
creating, by executing an instruction with the processor, a model based on the first value of the first feature, the model to establish a relationship between the first value of the first feature and the first emotion;
identifying, by executing an instruction with the processor, a second value of the first feature of first media evoking an unknown emotion;
identifying, by executing an instruction with the processor, the unknown emotion as the first emotion when the model indicates that the second value corresponds to the first value;
calculating, by executing an instruction with the processor, emotional distances between respective ones of potential media for recommendation and the first emotion; and
recommending, by executing an instruction with the processor, one of the potential media based on the respective emotional distances.