Singing Voice Edit Assistant Using Style Data

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing singing synthesizing technologies require manual adjustment of parameter values, making it difficult for users to easily adjust the individuality of a singing voice and add acoustic effects.

Innovation Solution

A singing voice edit assistant method that uses computer-adjusted score data and singing style data to automatically adjust the individuality and add acoustic effects, allowing users to specify music genres and tones for suitable synthesis.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If manual adjustment of parameter values is used, then the singing voice can be synthesized with precise control, but the ease of operation deteriorates because users need to manually adjust parameter values at each position

Engineering Contradiction:
Improveprecision of parameter adjustmentVSAvoidease of voice editing
Core Design Contradiction:
Measurement precisionVSEase of operation

Solution Approach 1:

The system performs preliminary analysis of the score data to automatically determine optimal parameter values before synthesis. The computer reads score data, analyzes musical structure, and pre-calculates parameter settings based on the music genre and phoneme characteristics, eliminating the need for manual parameter adjustment at each position while maintaining synthesis precision

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The system introduces an intermediate processing layer between the user input and the synthesis engine. This intermediary automatically translates high-level musical information (score data, genre, phonemes) into detailed parameter settings, serving as a mediator that converts simple user specifications into complex parameter configurations without requiring manual intervention

Inventive Principle:
Principle #24Intermediary (Mediator)

2Ease of operation

If automatic adjustment using score data and singing style data is used, then the ease of operation improves, but the manufacturing precision may deteriorate without proper parameter control

Engineering Contradiction:
Improveease of voice editingVSAvoidprecision of singing voice synthesis
Core Design Contradiction:
Ease of operationVSManufacturing precision

Solution Approach 1:

The system incorporates feedback mechanisms where the automatically adjusted parameters are evaluated against the score data and singing style data. The computer continuously refines parameter settings based on feedback from the synthesis results, ensuring that automatic adjustment maintains the precision required for high-quality singing voice synthesis while keeping operation simple

Inventive Principle:
Principle #23Feedback

Solution Approach 2:

The system dynamically changes parameters based on the analyzed score data and singing style data. Instead of using fixed parameters, the computer automatically adjusts parameters according to the specific musical context, genre requirements, and phoneme characteristics, maintaining synthesis precision through context-aware parameter optimization while preserving ease of operation

Inventive Principle:
Principle #35Parameter changes

Data Source

PatentEP3462442B1Singing voice edit assistant method and singing voice edit assistant device
Publication Date: 2020.09.09 YAMAHA CORP
  • EP3462442B1 patent drawingFigure 1~2
  • EP3462442B1 patent drawingFigure 3~4
  • EP3462442B1 patent drawingFigure 5A~6

AI summary

A singing voice edit assistant method, performed by a computer, includes: reading out singing style data that prescribes individuality of a singing voice and acoustic effects to be added to the singing voice, the singing voice being represented by singing voice data to be synthesized by the computer based on score data representing a time series of notes and lyrics data representing words corresponding to the respective notes; and synthesizing singing voice data while adjusting the individuality and adding acoustic effects based on the score data, the lyrics data, and the singing style data read out by the reading process.