Impression Formation Control via Listener Stimulation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for controlling impression formation in listeners through voice synthesis alter voice features, potentially misrepresenting the speaker's intention.
Innovation Solution
An impression formation control device that acquires a speaker's speech voice signal, extracts voice features, determines the bias in listener impression based on these features, and generates a stimulation control signal to adjust the listener's perception without altering the speaker's intention.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If voice synthesis technique is used to control impression formation, then impression formation control is achieved, but voice feature is altered causing speaker intention to be misrepresented
Solution Approach 1:
The system segments the voice signal processing into two independent paths: extracting acoustic features (pitch, timbre, rhythm) from the original speaker voice while separately synthesizing linguistic content. This allows selective manipulation of impression-related acoustic features without altering the core semantic message, thus maintaining speaker intention while achieving impression formation control.
Solution Approach 2:
The system introduces an intermediary processing layer that analyzes the original voice signal to extract acoustic characteristics, then uses these characteristics to guide synthetic voice generation. This intermediary process ensures that the synthesized voice maintains the essential intensional properties of the original speaker while allowing controlled modification of acoustic features that influence listener impression.
2Adaptability or versatility
If voice features are extracted and manipulated to control listener bias, then impression formation is controlled, but the original voice signal integrity is compromised
Solution Approach 1:
The system applies local quality modification by selectively altering specific acoustic features (such as pitch contour, spectral characteristics, or temporal patterns) that are known to influence listener impression, while preserving other aspects of the voice signal that carry the core speaker identity and intention. This localized manipulation achieves impression control without compromising overall voice signal integrity.
3Reliability
If synthesized voice is used to alter listener perception, then bias control is achieved, but naturalness of communication is reduced
Solution Approach 1:
The system changes specific acoustic parameters of the voice signal (such as fundamental frequency, formant structure, or speech rate) that are scientifically established to influence listener perception and bias. By carefully selecting and adjusting these parameters within natural ranges, the system achieves effective bias control while maintaining the naturalness and authenticity of the communication.
Data Source
AI summary
According to an embodiment of the present invention, when impression formation for a listener with respect to a speaker is controlled, a speech voice signal of the speaker is acquired, a voice feature is extracted from the speech voice signal, a bias to an impression made on the listener by the speech voice signal is determined based on the extracted voice feature, a bias control signal for controlling the bias occurs based on a determination result of the bias and information indicating a preset control direction of the bias, and a stimulation control signal for giving external stimulation to the listener is generated in accordance with the bias control signal, and the generated stimulation control signal is output.