Automatically acquiring feature segments in a music file

1. A method of automatically acquiring a feature segment in a music file, comprising:

receiving, with a processing device, a music file;

converting the music file into a character string;

evaluating at least one character string segment in the character string based on one or more music features comprising repetition times of a music segment, the evaluating comprising processing the character string to acquire a differential representation of the character string, and calculating repetition times of at least one character string segment in the differential representation; and

determining, based on an evaluation result, at least one music segment corresponding to at least one character string segment in the character string as a feature segment.

2. The method according to claim 1 , wherein the converting further comprises:

performing standard representation on the character string according to at least one of tone or syllable length.

3. The method according to claim 1 , wherein the evaluating further comprises:

evaluating at least one character string segment that exceeds a predetermined length in the character string.

4. The method according to claim 1 , wherein the music feature comprises an average pitch, and evaluating at least one character string segment in the character string based on one or more music features further comprises:

calculating an average pitch of a music segment corresponding to at least one character string segment in the character string.

5. The method according to claim 1 , wherein the music feature comprises segment location, and evaluating at least one character string segment in the character string based on one or more music features further comprises:

determining a location of at least one character string segment in the character string in the entire character string.

6. The method according to claim 1 , wherein the music feature comprises a segment length, and the evaluating at least one character string segment in the character string based on one or more music features further comprises:

calculating a segment length of at least one character string segment in the character string.

7. The method according to claim 1 , wherein the music feature comprises a semitone percentage, and evaluating at least one character string segment in the character string based on one or more music features further comprises:

calculating a semitone percentage of a music segment corresponding to at least one character string segment in the character string.

8. The method according to claim 1 , wherein determining at least one music segment as a feature segment further comprises:

determining a feature segment by using a classifier that has been trained with training data, wherein the training data comprise a given music file and its feature segment.

9. The method according to claim 1 , wherein determining at least one music segment as a feature segment further comprises:

determining a feature segment based on a result of comparing evaluation results of different character string segments in the character string.

10. The method according to claim 1 , wherein determining at least one music segment as a feature segment further comprises:

using a combination of a plurality of music segments corresponding to at least one character string segment in the character string as a feature segment.

11. A system for automatically acquiring a feature segment in a music file, comprising:

receiving means, configured to receive a music file;

converting means, configured to convert the music file into a character string;

evaluating means, configured to evaluate at least one character string segment in the character string based on one or more music features, comprising repetition times of a music segment, the evaluating means configured to process the character string to acquire a differential representation of the character string, and calculate repetition times of at least one character string segment in the differential representation; and

determining means, configured to determine, based on an evaluation result, at least one music segment corresponding to at least one character string segment in the character string as a feature segment.

12. The system according to claim 11 , wherein the converting means further comprises:

standardization means, configured to perform standard representation on the character string according to at least one of tone or syllable length.

13. The system according to claim 11 , wherein the evaluating means is further configured to evaluate at least one character string segment that exceeds a predetermined length in the character string.

14. The system according to claim 11 , wherein the music feature comprises an average pitch, and the evaluating means further comprises:

average pitch calculating means, configured to calculate an average pitch of a music segment corresponding to at least one character string segment in the character string.

15. The system according to claim 11 , wherein the music feature comprises segment location, and the evaluating means further comprises:

location determining means, configured to determine a location of at least one character string segment in the character string in the entire character string.

16. The system according to claim 11 , wherein the music feature comprises a segment length, and the evaluating means further comprises:

segment length calculating means, configured to calculate a segment length of at least one character string segment in the character string.

17. The system according to claim 11 , wherein the music feature comprises a semitone percentage, and the evaluating means further comprises:

percentage calculating means, configured to calculate a semitone percentage of a music segment corresponding to at least one character string segment in the character string.

18. The system according to claim 11 , wherein the determining means is further configured to determine a feature segment by using a classifier that has been trained with training data, wherein the training data comprises a music file and a specified feature segment.

19. The system according to claim 11 , wherein the determining means is further configured to determine a feature segment based on a result of comparing evaluation results of different character string segments in the character string.

20. The system according to claim 11 , wherein the determining means is further configured to use a combination of a plurality of music segments corresponding to at least one character string segment in the character string as a feature segment.