Music recommendation based on wearable devices

1 . A method for music recommendation using a wearable device, comprising:

obtaining one or more physiological parameters of a target user collected by the wearable device;

processing, by a processor, the one or more physiological parameters of the target user with a relaxation state assessment model to determine a current relaxation state of the target user;

determining, by the processor, at least one piece of target recommendation music based on at least one of the current relaxation state of the target user or relaxation parameters corresponding to multiple pieces of music to be recommended, wherein the at least one piece of target recommendation music is configured to be played for the target user, wherein the at least one piece of target recommendation music comprises multiple pieces of target recommendation music including a first music piece and a second music piece;

merging the multiple pieces of target recommendation music to obtain at least one piece of merged music, further comprising:

merging respective sub-segments, of a first music segment extracted from the first music piece and a second music segment extracted from the second music piece, based on a first weight sequence corresponding to the first music segment and a second weight sequence corresponding to the second music segment, wherein the first weight sequence and the second weight sequence each contain multiple weight values, the first weight sequence gradually decreasing in value and the second weight sequence gradually increasing in value; and

determining a playback order of the multiple pieces of target recommendation music and the at least one piece of merged music, wherein the multiple pieces of target recommendation music and the at least one piece of merged music are played for the target user in sequence based on the playback order.

2 . The method according to claim 1 , wherein the determining, by the processor, at least one piece of target recommendation music based on at least one of the current relaxation state of the target user or relaxation parameters corresponding to multiple pieces of music to be recommended comprises:

ranking the multiple pieces of music to be recommended in a descending order based on the relaxation parameters corresponding to the multiple pieces of music to be recommended;

determining at least one piece of target recommendation music from the multiple pieces of music to be recommended according to the ranking of the multiple pieces of music to be recommended.

3 . The method according to claim 1 , wherein the determining, by the processor, at least one piece of target recommendation music based on at least one of the current relaxation state of the target user or relaxation parameters corresponding to multiple pieces of music to be recommended comprises:

obtaining at least one previous relaxation state of the target user; and

determining at least one target recommendation music based at least in part on a difference of the current relaxation state and the at least one previous relaxation state of the target user.

4 . The method according to claim 1 , wherein the determining, by the processor, at least one piece of target recommendation music based on at least one of the current relaxation state of the target user or relaxation parameters corresponding to multiple pieces of music to be recommended comprises:

in response to determining that the target user is more relaxed based on the current relaxation state and at least one previous relaxation state, determining at least one piece of target recommendation music based at least in part on a piece of music currently being played for the target user, or

in response to determining that the target user is not more relaxed based on the current relaxation state and the at least one previous relaxation state, determining at least one piece of target recommendation music based on the relaxation parameters corresponding to the multiple pieces of music to be recommended.

5 . The method according to claim 1 , wherein the relaxation state assessment model is generated with a first training data set, and the first training data set comprises physiological parameters of multiple reference users and relaxation states annotated based on EEG data of the multiple reference users.

6 . The method according to claim 1 , further comprising:

before determining the at least one piece of target recommendation music based on at least one of the current relaxation state of the target user or relaxation parameters corresponding to multiple pieces of music to be recommended:

obtaining at least one of music preferences, attribute information, or historical sleep data of the target user;

determining a target music style corresponding to the target user based on at least one of the music preferences, attribute information, or historical sleep data of the target user; and

determining multiple pieces of candidate music belonging to the target music style from a candidate music library as the multiple pieces of music to be recommended.

7 . The method according to claim 1 , further comprising:

before determining the at least one piece of target recommendation music based on at least one of the current relaxation state of the target user or relaxation parameters corresponding to multiple pieces of music to be recommended:

obtaining music features of multiple pieces of music to be recommended; and

determining the relaxation parameters corresponding to the multiple pieces of music to be recommended with a relaxation parameter estimation model based on the music features of multiple pieces of music to be recommended and at least one of attribute information of the target user or historical music playback data of the target user.

8 . The method according to claim 7 , wherein the relaxation parameter estimation model is generated with a second training data set, and the second training data set comprises first relaxation curves obtained during playback of multiple pieces of reference music for multiple reference users and second relaxation curves obtained during the multiple pieces of reference music not being played for the multiple reference users.

9 . The method according to claim 1 , wherein the at least one piece of target recommendation music is one piece of target recommendation music; and

wherein the method further comprises:

after determining the at least one target recommendation music, merging the target recommendation music with a piece of music currently being played for the target user to obtain a piece of merged music, wherein the merged music is played for the target user before playing the target recommendation music.

10 . The method according to claim 1 , wherein merging the multiple pieces of target recommendation music to obtain at least one piece of merged music comprises:

determining, based on the relaxation parameters of the multiple pieces of target recommendation music, at least one pair of target recommendation music from the multiple pieces of target recommendation music, where each pair of target recommendation music comprises two pieces of adjacent target recommendation music; and

merging the two pieces of adjacent target recommendation music to obtain a piece of merged music;

wherein the merged music is played for the target user between corresponding two pieces of adjacent target recommendation music.

11 . The method according to claim 1 , wherein the second music piece is later in the playback order than the first music piece, and merging the multiple pieces of target recommendation music to obtain the at least one piece of merged music comprises:

extracting, starting from end, the first music segment of a first preset duration from the first music piece;

extracting, starting from beginning, the second music segment of a second preset duration from a latter one of the second music piece; and

merging the first music segment and the second music segment to obtain a corresponding one of the at least one merged music.

12 . The method according to claim 1 ,

wherein a sum of a first weight value from the first weight sequence and a corresponding second weight value from the second weight sequence is 1.

13 . A non-transitory computer-readable storage medium storing computer instructions causing a computer to execute the method according to claim 1 .

14 . A non-transitory computer program product comprising a computer program which, when executed by a computer, cause the computer to perform the method according to claim 1 .

15 . An electronic device, comprising:

at least one processor; and

a memory communicatively connected to the at least one processor; wherein,

the memory stores instructions executable by the at least one processor, and execution of the instructions by the at least one processor enables the at least one processor to:

obtain one or more physiological parameters of a target user collected by a wearable device;

process the one or more physiological parameters of the target user with a relaxation state assessment model to determine a current relaxation state of the target user;

determine at least one piece of target recommendation music based on at least one of the current relaxation state of the target user or relaxation parameters corresponding to multiple pieces of music to be recommended, wherein the at least one piece of target recommendation music comprises multiple pieces of target recommendation music including a first music piece and a second music piece, and the at least one piece of target recommendation music is configured to be played for the target user,

merge the multiple pieces of target recommendation music to obtain at least one piece of merged music, wherein the instructions further comprise instructions to:

merge respective sub-segments, of a first music segment extracted from the first music piece and a second music segment extracted from the second music piece, based on a first weight sequence corresponding to the first music segment and a second weight sequence corresponding to the second music segment, wherein the first weight sequence and the second weight sequence each contain multiple weight values, the first weight sequence gradually decreasing in value and the second weight sequence gradually increasing in value; and

determine a playback order of the multiple pieces of target recommendation music and the at least one piece of merged music, wherein the multiple pieces of target recommendation music and the at least one piece of merged music are played for the target user in sequence based on the playback order.

16 . The electronic device according to claim 15 , wherein

the electronic device is the wearable device, a mobile terminal wirelessly communicating with the wearable device, or a remote server wirelessly communicating with the mobile terminal.

17 . The electronic device according to claim 15 , wherein

the electronic device communicates wirelessly with the wearable device in a short range.

18 . The electronic device according to claim 15 , wherein the at least one process is further enabled to:

before determining the at least one piece of target recommendation music based on at least one of the current relaxation state of the target user or relaxation parameters corresponding to multiple pieces of music to be recommended:

obtain at least one of music preferences, attribute information, or historical sleep data of the target user;

determine a target music style corresponding to the target user based on at least one of the music preferences, attribute information, or historical sleep data of the target user; and

determine multiple pieces of candidate music belonging to the target music style from a candidate music library as the multiple pieces of music to be recommended.

19 . The electronic device according to claim 15 , wherein the at least one process is further enabled to:

before determining the at least one piece of target recommendation music based on at least one of the current relaxation state of the target user or relaxation parameters corresponding to multiple pieces of music to be recommended:

obtain music features of multiple pieces of music to be recommended; and

determine the relaxation parameters corresponding to the multiple pieces of music to be recommended with a relaxation parameter estimation model based on the music features of multiple pieces of music to be recommended and at least one of attribute information or historical music playback data of the target user.