Method and apparatus for storing/searching for music

• US 9,012,755 B2
• Filed: 04/15/2008
• Issued: 04/21/2015
• Est. Priority Date: 01/07/2008
• Status: Active Grant

• Current Assignee: Huawei Technologies Co., Ltd. (Huawei Investment & Holding Co., Ltd.)
• Original Assignee: Samsung Electronics Co. Ltd.
• Inventors: Eom, Ki-wan, Lee, Jae-won, Kim, Jung-eun, Deng, Jing
• Primary Examiner(s): FLETCHER, MARLON T
• Application Number:
  US12/081,410
• Publication Number:
  US 20090173214A1
• Time in Patent Office: 2,562 Days
• Field of Search: None
• US Class Current: 84/616
• CPC Class Codes: G06F 16/634 Query by example, e.g. query by humming
• CPC Class Codes: G06F 16/685 using automatically derived transcript of audio data, e.g. lyrics speech recognition G10L15/00
• CPC Class Codes: G10H 1/0008 Associated control or indicating means
• CPC Class Codes: G10H 2210/061 for extraction of musical phrases, isolation of musically relevant segments, e.g. musical thumbnail generation, or for temporal structure analysis of a musical piece, e.g. determination of the movement sequence of a musical work
• CPC Class Codes: G10H 2210/071 for rhythm pattern analysis or rhythm style recognition
• CPC Class Codes: G10H 2220/011 Lyrics displays, e.g. for karaoke applications
• CPC Class Codes: G10H 2240/141 Library retrieval matching, i.e. any of the steps of matching an inputted segment or phrase with musical database contents, e.g. query by humming, singing or playing; the steps may include, e.g. musical analysis of the input, musical feature extraction, query formulation, or details of the retrieval process

Abstract

Provided are a method and apparatus for generating and storing a database that may be searched for music upon a user'"'"'s request, and a method and apparatus for searching for music corresponding to the user'"'"'s request. According to an embodiment of the present invention, a plurality of melodies are divided based on phrase breaks and are stored, and a number of melodies which are similar to the user'"'"'s request are detected from the stored melodies.

Claims

1. An apparatus for searching music, the apparatus comprising:

an input unit to receive a user'"'"'s request including sound generated by at least one of the user and a device;

a melody extraction unit to extract a melody from the sound; and

a melody detection unit to detect music corresponding to the extracted melody by comparing the extracted melody with the melodies stored in a melody storage unit, based on phrase breaks of lyrics of the stored melodies and output information of the detected music,

wherein the phrase breaks of the lyrics are predicted using at least one of a syntactic structure of a sentence, information on parts of speech, and phrase location information.

2. The apparatus of claim 1, wherein the melody storage unit divides the melodies based on note lengths.

3. The apparatus of claim 1, further comprising:

a lyrics storage unit for dividing lyrics of the melodies so as to correspond to the melodies, and storing the lyrics; and

a lyrics detection unit for detecting a number of lyrics which are similar to the user'"'"'s request, from the stored lyrics, in correspondence with the number of melodies detected by the melody detection unit.

4. The apparatus of claim 3, wherein the lyrics detection unit detects the number of melodies by recognizing speech from the user'"'"'s request and comparing the speech to words included only in the number of lyrics detected by the lyrics detection unit.

5. The apparatus of claim 1, wherein the melody storage unit comprises:

a first melody storage unit for storing melodies considered to be frequently requested by users; and

a second melody storage unit for storing melodies considered to be rarely requested by users.

6. The apparatus of claim 5, wherein the first melody storage unit comprises:

a first phrase melody storage unit for storing first phrase melodies of songs;

a repeated melody storage unit for storing repeated melodies of the songs which are repeated more than a preset number of times; and

a chorus melody storage unit for storing chorus melodies of the songs.

7. The apparatus of claim 5, wherein the melody detection unit searches the first melody storage unit and then searches the second melody storage unit.

8. The apparatus of claim 3, wherein the lyrics storage unit comprises:

a first lyrics storage unit for storing lyrics considered to be frequently requested by users or lyrics corresponding to melodies considered to be frequently requested by users; and

a second lyrics storage unit for storing lyrics considered to be rarely requested by users or lyrics corresponding to melodies considered to be rarely requested by users.

9. The apparatus of claim 8, wherein the first lyrics storage unit comprises:

a first phrase lyrics storage unit for storing lyrics corresponding to first phrase melodies of the songs;

a repeated melody lyrics storage unit for storing lyrics corresponding to repeated melodies of the songs which are repeated more than a preset number of times; and

a chorus lyrics storage unit for storing lyrics corresponding to chorus melodies of the songs.

10. The apparatus of claim 8, wherein the lyrics detection unit searches the first lyrics storage unit and then searches the second lyrics storage unit.

11. The apparatus of claim 1, wherein the phrase breaks of the lyrics are predicted using a prosodic break index method that comprises a hidden Markov model (HMM) method, a neural network (NN) method, and a classification and regression tree (CART) method.

12. An apparatus to store music, the apparatus comprising:

a bar extraction unit to receive melodies and extract bars of the melodies;

a phrase break prediction unit to predict phrase breaks of the lyrics;

a division unit to divide melodies based on the bars extracted by the bar extraction unit and the phrase breaks predicted by the phrase break prediction unit; and

a melody storage unit to store the divided melodies,

wherein the phrase breaks of the lyrics are predicted using at least one of a syntactic structure of a sentence, information on parts of speech, and phrase location information.

13. The apparatus of claim 12, wherein the division unit divides the melodies based on note lengths.

14. The apparatus of claim 12, wherein the division unit does not divide the melodies at a bar included in a phrase.

15. The apparatus of claim 12, wherein the division unit divides the melodies at an end of a phrase predicted to be an intonational phrase (IP).

16. The apparatus of claim 12, wherein the division unit determines whether to divide the melodies at an end of a phrase predicted to be an accentual phrase (AP), in accordance with a length of a last note of the phrase.

17. The apparatus of claim 13, wherein the division unit determines to divide the melodies at the end of the phrase if the length of the last note of the phrase is longer than a preset length.

18. The apparatus of claim 12, further comprising a lyrics storage unit for dividing lyrics of the melodies so as to correspond to the melodies, and storing the lyrics.

19. The apparatus of claim 12, further comprising a classification unit for classifying the melodies into melodies considered to be frequently requested by users and melodies considered to be rarely requested by users, and

wherein the melody storage unit comprises:

a first melody storage unit for storing the melodies considered to be frequently requested by users; and

a second melody storage unit for storing the melodies considered to be rarely requested by users.

20. The apparatus of claim 19, wherein the first melody storage unit classifies and separately stores at least one of first phrase melodies of songs, repeated melodies of the songs which are repeated more than a preset number of times, and chorus melodies of the songs.

21. The apparatus of claim 18, wherein the lyrics storage unit comprises:

a first lyrics storage unit for storing lyrics considered to be frequently requested by users or lyrics corresponding to melodies considered to be frequently requested by users; and

a second lyrics storage unit for storing lyrics considered to be rarely requested by users or lyrics corresponding to melodies considered to be rarely requested by users.

22. The apparatus of claim 21, wherein the first lyrics storage unit classifies and separately stores at least one of lyrics corresponding to first phrase melodies of the songs, lyrics corresponding to repeated melodies of the songs which are repeated more than a preset number of times, and lyrics corresponding to chorus melodies of the songs.

23. A method of searching for music, comprising:

receiving a request of a sound to be recognized;

extracting a melody of the request;

detecting a preset number of melodies similar to the extracted melody;

recognizing only a speech of the request, if the preset number of melodies have a similarity value equal to or less than a preset threshold value;

detecting a preset number of lyrics similar to the recognized speech, if the preset number of melodies have a similarity value greater than a preset threshold value; and

outputting information corresponding to the detecting results.