Methods and systems for identifying similar songs

• US 7,812,241 B2
• Filed: 09/27/2007
• Issued: 10/12/2010
• Est. Priority Date: 09/27/2006
• Status: Expired due to Fees

• Current Assignee: Trustees Of Columbia University In The City Of New York (Columbia University)
• Original Assignee: Trustees Of Columbia University In The City Of New York (Columbia University)
• Inventors: Ellis, Daniel P. W.
• Primary Examiner(s): Donels; Jeffrey
• Application Number:
  US11/863,014
• Publication Number:
  US 20080072741A1
• Time in Patent Office: 1,111 Days
• Field of Search: 846/12, 846/36
• US Class Current: 84/612
• CPC Class Codes: G10H 1/0008 Associated control or indicating means
• CPC Class Codes: G10H 1/40 Rhythm
• CPC Class Codes: G10H 2210/076 for extraction of timing, tempo; Beat detection tempo display G10H2220/081; tempo control G10H2210/375
• CPC Class Codes: G10H 2240/135 Library retrieval index, i.e. using an indexing scheme to efficiently retrieve a music piece
• 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
• CPC Class Codes: G10H 2250/135 Autocorrelation

Abstract

Methods and systems for identifying similar songs are provided. In accordance with some embodiments, methods for identifying similar songs are provided, the methods comprising: identifying beats in at least a portion of a song; generating beat-level descriptors of the at least a portion of the song corresponding to the beats; comparing the beat-level descriptors to other beat-level descriptors corresponding to a plurality of songs. In accordance with some embodiments, systems for identifying similar songs are provided, the systems comprising: a digital processing device that: identifies beats in at least a portion of a song; generates beat-level descriptors of the at least a portion of the song corresponding to the beats; and compares the beat-level descriptors to other beat-level descriptors corresponding to a plurality of songs.

Claims

1. A method for comparing songs, comprising:

receiving at least a portion of a song using a microphone in a mobile telephone; and

using a microprocessor in the mobile telephone;

identifying beat times for beats in the at least a portion of the song;

generating beat-level descriptors of the at least a portion of the song corresponding to the beats;

comparing the beat-level descriptors to other beat-level descriptors corresponding to a plurality of songs; and

identifying one of the plurality of songs on a display of the mobile telephone based on the comparing,

wherein identifying the beat times for the beats comprises forming an onset strength envelope for the at least a portion of the song, determining a primary tempo period estimate, identifying a beat in the beats, and back tracking from the beat to earlier-occurring beats.

2. A method for comparing songs, comprising:

receiving at least a portion of a song using a microphone in a mobile telephone; and

using a microprocessor in the mobile telephone;

identifying beat times for beats in the at least a portion of the song;

generating beat-level descriptors of the at least a portion of the song corresponding to the beats;

comparing the beat-level descriptors to other beat-level descriptors corresponding to a plurality of songs; and

identifying one of the plurality of songs on a display of the mobile telephone based on the comparing,

wherein generating beat-level descriptors comprises generating chroma bins for each beat of the portion of the song.

3. The method of claim 2, wherein the chroma bins are generated using a Fourier transform.

4. The method of claim 2, wherein the chroma bins span one octave, and further comprising mapping a plurality of octaves in the portion of the song to the same chroma bins.

5. A method for comparing songs, comprising:

receiving at least a portion of a song using a microphone in a mobile telephone; and

using a microprocessor in the mobile telephone;

identifying beat times for beats in the at least a portion of the song;

generating beat-level descriptors of the at least a portion of the song corresponding to the beats;

comparing the beat-level descriptors to other beat-level descriptors corresponding to a plurality of songs; and

identifying one of the plurality of songs on a display of the mobile telephone based on the comparing,

wherein comparing the beat-level descriptors to other beat-level descriptors comprises performing a cross-correlation on the beat-level descriptors.

6. The method of claim 5, wherein the cross-correlation comprises performing a Fourier transform on the beat-level descriptors.

7. A method for comparing songs, comprising:

receiving at least a portion of a song using a microphone in a mobile telephone; and

using a microprocessor in the mobile telephone;

identifying beat times for beats in the at least a portion of the song;

generating beat-level descriptors of the at least a portion of the song corresponding to the beats;

comparing the beat-level descriptors to other beat-level descriptors corresponding to a plurality of songs; and

identifying one of the plurality of songs on a display of the mobile telephone based on the comparing,

wherein comparing the beat-level descriptors to other beat-level descriptors comprises identifying boundaries in the beat-level descriptors and performing a nearest neighbor search.

8. The method of claim 7, wherein the nearest neighbor search is a Locality-Sensitive Hash.

9. A method for comparing songs, comprising:

receiving at least a portion of a song using a microphone in a mobile telephone;

using a microprocessor in the mobile telephone;

identifying beat times for beats in the at least a portion of the song;

generating beat-level descriptors of the at least a portion of the song corresponding to the beats;

comparing the beat-level descriptors to other beat-level descriptors corresponding to a plurality of songs; and

identifying one of the plurality of songs on a display of the mobile telephone based on the comparing; and

identifying at least one of the song and the plurality of songs as a cover song of another of the at least one of the song and the plurality of songs.

10. A device for comparing songs, comprising:

a mobile telephone including;

a microphone that receives at least a portion of a song;

a microprocessor that;

identifies beat times for beats in the at least a portion of the song;

generates beat-level descriptors of the at least a portion of the song corresponding to the beats; and

compares the beat-level descriptors to other beat-level descriptors corresponding to a plurality of songs; and

a display that identifies one of the plurality of songs based on the comparing,

wherein the microprocessor, in identifying the beat times for the beats, also forms an onset strength envelope for the at least a portion of the song, determines a primary tempo period estimate, identifies a beat in the beats, and back tracks from the beat to earlier-occurring beats.

11. A device for comparing songs, comprising:

a mobile telephone including;

a microphone that receives at least a portion of a song;

a microprocessor that;

identifies beat times for beats in the at least a portion of the song;

generates beat-level descriptors of the at least a portion of the song corresponding to the beats; and

compares the beat-level descriptors to other beat-level descriptors corresponding to a plurality of songs; and

a display that identifies one of the plurality of songs based on the comparing,

wherein the microprocessor, in generating beat-level descriptors, also generates chroma bins for each beat of the portion of the song.

12. The device of claim 11, wherein the chroma bins are generated using a Fourier transform.

13. The device of claim 11, wherein the chroma bins span one octave, and the microprocessor also maps a plurality of octaves in the portion of the song to the same chroma bins.

14. A device for comparing songs, comprising:

a mobile telephone including;

a microphone that receives at least a portion of a song;

a microprocessor that;

identifies beat times for beats in the at least a portion of the song;

generates beat-level descriptors of the at least a portion of the song corresponding to the beats; and

compares the beat-level descriptors to other beat-level descriptors corresponding to a plurality of songs; and

a display that identifies one of the plurality of songs based on the comparing,

wherein the microprocessor, in comparing the beat-level descriptors to other beat-level descriptors, also performs a cross-correlation on the beat-level descriptors.

15. The device of claim 14, wherein the microprocessor, in performing the cross-correlation, also performs a Fourier transform on the beat-level descriptors.

16. A device for comparing songs, comprising:

a mobile telephone including;

a microphone that receives at least a portion of a song;

a microprocessor that;

identifies beat times for beats in the at least a portion of the song;

generates beat-level descriptors of the at least a portion of the song corresponding to the beats; and

compares the beat-level descriptors to other beat-level descriptors corresponding to a plurality of songs; and

a display that identifies one of the plurality of songs based on the comparing,

wherein the microprocessor, in comparing the beat-level descriptors to other beat-level descriptors, also identifies boundaries in the beat-level descriptors and performs a nearest neighbor search.

17. A device for comparing songs, comprising:

a mobile telephone including;

a microphone that receives at least a portion of a song;

a microprocessor that;

identifies beat times for beats in the at least a portion of the song;

generates beat-level descriptors of the at least a portion of the song corresponding to the beats; and

compares the beat-level descriptors to other beat-level descriptors corresponding to a plurality of songs; and

a display that identifies one of the plurality of songs based on the comparing,

wherein the display also identifies at least one of the song and the plurality of songs as a cover song of another of the at least one of the song and the plurality of songs.