Emotion Intensity and its Control for Emotional Voice Conversion

TL;DR

This paper introduces a method for emotional voice conversion that explicitly models and controls emotion intensity, enabling more expressive and nuanced speech synthesis while preserving content and speaker identity.

Contribution

It proposes a novel approach to disentangle speaker style from content and encode emotion intensity in a continuous space, improving emotional expressiveness in voice conversion.

Findings

Effective control of emotion intensity demonstrated

Improved emotional expressiveness validated through evaluations

Disentanglement of style and content enhances conversion quality

Abstract

Emotional voice conversion (EVC) seeks to convert the emotional state of an utterance while preserving the linguistic content and speaker identity. In EVC, emotions are usually treated as discrete categories overlooking the fact that speech also conveys emotions with various intensity levels that the listener can perceive. In this paper, we aim to explicitly characterize and control the intensity of emotion. We propose to disentangle the speaker style from linguistic content and encode the speaker style into a style embedding in a continuous space that forms the prototype of emotion embedding. We further learn the actual emotion encoder from an emotion-labelled database and study the use of relative attributes to represent fine-grained emotion intensity. To ensure emotional intelligibility, we incorporate emotion classification loss and emotion embedding similarity loss into the training of the EVC network. As desired, the proposed network controls the fine-grained emotion intensity in the output speech. Through both objective and subjective evaluations, we validate the effectiveness of the proposed network for emotional expressiveness and emotion intensity control.