Melody Encoding System Using Relative Pitch and Rhythm Codes
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current Music Information Retrieval (MIR) systems are inefficient and require extensive musical training to locate songs using melodic content, as they often yield inconsistent results and are difficult to use, especially when users can only remember excerpts of musical content like melodies without textual information.
Innovation Solution
A system and method that encodes melodic information using relative pitch and rhythm changes, employing Parsons and Kolta codes, and matches these encodings with a Smith-Waterman algorithm to retrieve musical data from a database, allowing users to input melodies through singing, humming, or playing, without requiring formal musical training.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If traditional MIR methods are used to retrieve music using melodic content, then music retrieval functionality is provided, but the system requires extensive musical training and yields inconsistent results
Solution Approach 1:
The patent introduces an intermediary encoding system (Parsons code and Kolta code) that translates melodic input into a standardized format. This intermediary layer allows users without musical training to input melodies by singing or humming, which are then converted into comparable encoded representations for database searching, resolving the contradiction between reliability and ease of operation.
Solution Approach 2:
The patent transforms the parameter representation of melodic data from absolute pitch values to relative interval changes (Parsons code) and relative duration ratios (Kolta code). This parameter transformation makes the representation more robust to variations in singing pitch and tempo, improving query effectiveness while allowing non-musical users to input melodies flexibly.
2Measurement precision
If detailed melodic information is required for accurate song identification, then retrieval accuracy improves, but the complexity of the system increases
Solution Approach 1:
The patent extracts only the essential features of melodic content—relative pitch intervals and relative duration ratios—while discarding non-essential information such as absolute pitch, tempo, and articulation details. This extraction maintains sufficient accuracy for song identification while significantly simplifying the encoding and matching processes.
Solution Approach 2:
The patent segments the melodic input into discrete note events, each characterized by relative pitch interval and relative duration. This segmentation allows for systematic encoding using Parsons and Kolta codes, and enables efficient comparison using the Smith-Waterman algorithm, balancing accuracy with computational simplicity.
3Ease of operation
If a fragment of melody is used for query instead of complete song, then user convenience improves, but the ability to uniquely identify the song decreases
Solution Approach 1:
The patent applies partial action by allowing users to input only a fragment of the melody rather than the complete song. The relative interval and duration encoding, combined with the Smith-Waterman similarity search, enables accurate identification even from partial inputs, as the algorithm can match patterns against the database despite the incomplete nature of the query.
Data Source
AI summary
A system and method for retrieving musical information from a database based on a melody fragment. The system may be used by a person without formal musical training to retrieve musical data (e.g., the name of a song, bibliographic information about a song, or the song itself) from a database by providing the melody or fragment of the melody of the desired music to a computer interface. The melody may be provided by singing, humming, whistling, or playing a musical instrument, for example, a keyboard. The inputted melodic information is encoded using relative changes in pitch and rhythm throughout the melody. The encoded information is then compared using a matching algorithm to similarly encoded melodic information representative of many musical pieces (e.g., songs). Results may also be sorted using a weighted algorithm.