Back to results
US 2014/0174279 A1Published application
Composition using correlation between melody and lyrics
AssigneeThe Hong Kong University of Science and Technology
InventorsChi Wing WONG, Raymond Ka Wai SZE, Cheng LONG
Life of this patent
Estimating the term…
Abstract
Disclosed are ways to generate a melody. Currently, no algorithm exists for automatically composing a melody based on music lyrics. However, according to some recent studies, within a song, there usually exists a correlation between a song's notes and a song's lyrics wherein a melody can be generated based on such correlation. Disclosed herein, are systems, methods and algorithms that consider the correlation between a song's lyrics and a song's notes to compose a melody.
Claims22
A method, comprising:
- receiving, by a system comprising a processor from a data store, tone data determined from a set of songs represented by a set of notes and a set of song lyrics represented by a set of words, wherein the tone data is selected from the data store based at least on first correlation data that correlates the set of notes to the set of words;
- determining, by the system, a pattern at least based on a correlation between a subset of the songs represented by a subset of the notes and a subset of the song lyrics represented by a subset of the words;
- creating, by the system, a composition model based at least on the pattern;
- generating, by the system, a melody based at least on the composition model;
- and
- pairing, by the system, the melody at least to the subset of the song lyrics.
3 more elements
A system, comprising:
- a processor, coupled to a memory, that executes or facilitates execution of one or more executable components, comprising:
- a tone extraction component that selects tone data from a set of songs associated with a set of notes and a set of song lyrics represented by a set of words, from a data store, wherein the selection is based at least on first correlation data representing a correlation between the set of notes and the set of words;
- a pattern mining component that determines a pattern at least based on second correlation data representing a correlation between a subset of notes and a subset of words associated with respective songs of the set of songs;
- an automatic modeling component that creates an automatic composition model at least based on the pattern;
- and
- a generation component that generates a melody at least based on the automatic composition model.
3 more elements
A system, comprising:
- a processor, coupled to a memory, that executes or facilitates execution of executable instructions to at least:
- generate a melody based on first correlation data that represents a correlation between note data and word data;
- convert the word data into wave data;
- translate the wave data into vocal data;
- and
- simulate a human singing a song based on the vocal data and a melody generated from the first correlation data.
3 more elements
Dependencies are inferred from the claim text.