Dynamic Character Modification for Voice Synthesis Diversity

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Existing voice synthesis techniques generate monotonous voices when vocalizing pre-prepared lyrics, and preparing diverse lyrics to overcome this issue incurs a significant workload.

Innovation Solution

A voice synthesis method that dynamically changes character strings for vocalization based on predetermined conditions, using a computer system with a synthesis manager and voice synthesizer to generate sound signals, allowing for rich vocal content without excessive workload.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Adaptability or versatility

If numerous sets of different lyrics are prepared in advance to overcome monotony, then vocal diversity is improved, but work load increases excessively

Engineering Contradiction:

Improvevocal diversityVSAvoidwork load

Core Design Contradiction:

Adaptability or versatilityVSDevice complexity

Solution Approach 1:

The system dynamically changes character strings for vocalization based on information processing conditions rather than relying on pre-prepared diverse lyrics. The synthesis manager modifies characters in real-time during voice generation, enabling vocal diversity without the need to maintain multiple lyric sets in advance.

Inventive Principle:

Principle #15Dynamics

Solution Approach 2:

The voice synthesis system performs self-modification of character strings through automatic information processing. The synthesis manager autonomously determines whether to change characters based on processing conditions, eliminating the need for external preparation of multiple lyric versions and reducing manual workload.

Inventive Principle:

Principle #25Self-service

2Productivity

If pre-prepared lyrics are used for voice synthesis, then generation speed is improved, but vocal content richness deteriorates

Engineering Contradiction:

Improvegeneration speedVSAvoidvocal content richness

Core Design Contradiction:

ProductivityVSAdaptability or versatility

Solution Approach 1:

The system introduces dynamic character string modification during the voice generation process. Based on information processing conditions, the synthesis manager selectively changes characters to enhance vocal content richness while maintaining efficient generation speed through automated decision-making rather than manual intervention.

Inventive Principle:

Principle #15Dynamics

Solution Approach 2:

The system changes parameters of character strings (such as character selection, modification timing, and change content) based on information processing conditions. This allows the voice synthesis to adapt vocal content richness dynamically without sacrificing generation efficiency, as the changes are applied programmatically rather than requiring manual lyric preparation.

Inventive Principle:

Principle #35Parameter changes

Data Source

PatentUS10204617B2Voice synthesis method and voice synthesis device

Publication Date: 2019.02.12 YAMAHA CORP

AI summary

An information processing device determines whether a predetermined condition with regard to information processing has been met, changes a character for vocalization when the predetermined condition has been met, and generates a sound signal of a synthesized voice obtained by vocalizing the character for vocalization that has been changed.