Emotion Intensity and its Control for Emotional Voice Conversion

Kun Zhou, Berrak Sisman, Rajib Rana, Bjorn W. Schuller, Haizhou Li

Research output: Contribution to journalArticlepeer-review

32 Scopus citations

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.

Original languageEnglish
Pages (from-to)31-48
Number of pages18
JournalIEEE Transactions on Affective Computing
Volume14
Issue number1
DOIs
StatePublished - 1 Jan 2023
Externally publishedYes

Keywords

  • Emotional voice conversion
  • emotion intensity
  • limited data
  • perceptual loss
  • relative attribute
  • sequence-to-sequence

Fingerprint

Dive into the research topics of 'Emotion Intensity and its Control for Emotional Voice Conversion'. Together they form a unique fingerprint.

Cite this