SYSTEM AND METHOD FOR ADAPTIVE MELODIC SEGMENTATION AND MOTIVIC IDENTIFICATION

• US 20100251876A1
• Filed: 05/11/2010
• Published: 10/07/2010
• Est. Priority Date: 12/31/2007
• Status: Active Grant

• Current Assignee: Orpheus Media Research, LLC
• Original Assignee: Orpheus Media Research, LLC
• Inventors: WILDER, Gregory W.
• Granted Patent: US 8,084,677 B2
• Time in Patent Office: Days
• US Class Current: 84/609
• CPC Class Codes: G10H 1/0008 Associated control or indicating means
• CPC Class Codes: G10H 2210/066 for pitch analysis as part of wider processing for musical purposes, e.g. transcription, musical performance evaluation; Pitch recognition, e.g. in polyphonic sounds; Estimation or use of missing fundamental
• 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 2210/086 for transcription of raw audio or music data to a displayed or printed staff representation or to displayable MIDI-like note-oriented data, e.g. in pianoroll format
• CPC Class Codes: G10H 3/125 Extracting or recognising the pitch or fundamental frequency of the picked up signal

Abstract

The present invention comprises a system and method, modeled on research observations in human perception and cognition, capable of accurately segmenting primarily (although not exclusively) melodic input in performance data and encoded digital audio data, and mining the results for defining motives within the input data.

Claims

1. In a computer system, a method for characterizing a data set representative of music comprising:

a. identifying within the data set at least one subset of data representing melody;

b. identifying within said subset of data representing melody at least one subset of data representing at least one motive.

2. The method of claim 1 wherein the data set representative of music comprises encoded digital audio.

3. The method of claim 1 wherein the data set representative of music comprises encoded performance information.

4. The method of claim 1 wherein the data set representative of music comprises both encoded performance information and encoded digital audio.

5. The method of claim 1 wherein the data set representative of music comprises information representative of one or more of the following characteristics:

phrase structure, measure information, tempo information, section identifiers, stylistic attributes, exact pitch, onset, offset, velocity, and note density.

6. The method of claim 1 wherein the motive is identified using adaptive melodic segmentation.

7. The method of claim 1 wherein the step of identifying the subset of data representing melody comprises:

a. identifying note events within the data set;

b. determining at least one attribute for each note event, the attributes being selected from among pitch, onset, length and velocity; and

c. determining the attributes of difference between consecutive note events.

8. The method of claim 7 wherein the step of identifying note events is optionally preceded by at least one preprocessing step selected from the group of:

equal loudness contour filtering, spectral pitch tracking, tempo extraction, instantaneous frequency distribution analysis, and style identification.

9. The method of claim 7 wherein the step of identify the subset of data representing melody additionally comprises:

a. determining pitch contour between each note event pair and determining note length for each note event.

10. The method of claim 9 further comprising the step of segmenting the subset of data representing melody into motive segments.

11. method of claim 10 wherein the segmentation is performed by a method comprising:

a. defining thresholds of change of the attributes or contours;

b. weighting the defined thresholds; and

c. determining boundaries within the subset of data representing melody;

d. identifying motive segments between the boundaries.

12. The method of claim 11 wherein consecutive motive segments are combined into a single, larger motive segment by similarity ballooning.

13. A computer-implemented system for characterizing a data set representative of music comprising:

a. means for identifying within the data set at least one subset of data representing melody; and

b. means for identifying within said subset of data representing melody at least one subset of data representing at least one motive.

14. A method for quantitatively assessing the similarities between or among a plurality of data sets representative of music wherein one or more accused data sets represents an alleged infringement of copyright in one or more proprietary data sets comprising the steps of:

a. Processing each of the data sets to identify the motives therein by adaptive melodic segmentation; and

b. Comparing the motives identified to establish the degree of similarity between or among the data sets.