The Emotional Voices Database: Towards Controlling the Emotion Dimension in Voice Generation Systems

TL;DR

This paper introduces an open-source emotional speech database designed for voice synthesis, demonstrating its effectiveness through a simple conversion system that controls emotional expression in speech.

Contribution

The paper provides a new, publicly available emotional speech database covering multiple languages and emotions, enabling better control in voice generation systems.

Findings

The database effectively supports emotion-controlled speech synthesis.

A simple MLP system successfully converts neutral to angry speech.

Perception tests confirm the database's usefulness for future research.

Abstract

In this paper, we present a database of emotional speech intended to be open-sourced and used for synthesis and generation purpose. It contains data for male and female actors in English and a male actor in French. The database covers 5 emotion classes so it could be suitable to build synthesis and voice transformation systems with the potential to control the emotional dimension in a continuous way. We show the data's efficiency by building a simple MLP system converting neutral to angry speech style and evaluate it via a CMOS perception test. Even though the system is a very simple one, the test show the efficiency of the data which is promising for future work.