Application US20190246936

SYSTEM AND METHOD FOR ASSOCIATING MUSIC WITH BRAIN-STATE DATA
[+] Show Details

US20190246936

1. A music system comprising: (10) (10)

(a) at least one bio-signal sensor configured to capture bio-signal sensor data from at least one user;

(b) an input receiver configured to receive music data and the bio-signal sensor data, the music data and the bio-signal sensor data being temporally defined such that the music data corresponds temporally to at least a portion of the bio-signal sensor data;

(i) a music processor to segment the music data into a plurality of time epochs of music, each epoch of music linked to a time stamp;

(ii) a sonic feature extractor to, for each epoch of music, extract a set of sonic features;

(iii) a biological feature extractor to extract, for each epoch of music, a set of biological features from the bio-signal sensor data using the time stamp for the respective epoch of music;

(iv) a metadata extractor to extract metadata from the music data;

(v) a user feature extractor to extract a set of user attributes from the music data and the bio-signal sensor data;

(vi) a machine learning engine to transform the set of sonic features, the set of biological features, the set of metadata, and the set of user attributes into, for each epoch of music, a set of categories that the respective epoch belongs to using one or more predictive models to predict a user reaction to music; and

(d) a music recommendation engine configured to provide at least one music recommendation based on the sets of categories.

2. The music system of claim 1 wherein the input receiver is configured to receive a target emotional state and wherein the system further comprises a music controller to interact with the music recommendation engine to provide at least one music recommendation based on the target emotional state. (0)

3. The music system of claim 1 wherein the music processor is configured to identify a selection of songs from the music data and add a temporal history to the selection of songs, the temporal history indicating a date and time a user of the at least one users listened to or otherwise selected songs of the selection of songs and an order of the selection of songs, wherein the predictive models use a temporal model for the temporal history the selection of songs. (1) (0)

5. The music system of claim 1 wherein the at least one processor is configured to provide the one or more predictive models has a learning mode and an operational mode. (0)

6. The music system of claim 1 wherein each of the categories are nested in a hierarchy of nodes or an ordered list of probabilities of the respective category. (0)

7. The music system of claim 1 further comprising a plurality of bio-signal sensors configured to capture bio-signal sensor data from a plurality of users and correlate a portion of the bio-signal sensor data across the plurality of users, the portion being mapped to one or more epochs of music, wherein the machine learning engine transform the portion of the bio-signal sensor data across the plurality of users to a common set of categories for use in music recommendations. (0)

8. The music system of claim 1 further comprising a music effect controller to influence user state by playback or recording of music based the at least one music recommendation. (0)

9. The music system of claim 1 wherein the machine learning engine is to further transform a set of user actions taken during playback of the music data into a set of categories. (1) (0)

11. The music system of claim 1 wherein the input receiver includes a microphone to receive music data. (1) (0)

15. The music system of claim 1 wherein the at least one user is a plurality of users, the music data includes instances of a music track associated with each of the plurality of users, each instance of the music track having standardized tags, and the plurality of time epochs of music for each instance are identified based at least in part on the standardized tags. (1) (0)

17. The music system of claim 1 wherein the at least one user is a plurality of users, and each of the plurality of users listens to a music track concurrently. (1) (0)

View Abstract and Specification Size

PDF with Images and Document Face >

Full Text Publication >

Application US20190246936

SYSTEM AND METHOD FOR ASSOCIATING MUSIC WITH BRAIN-STATE DATA

Patent Matrix® Search