Voice Conversion for Likability Control via Automated Rating of Speech Synthesis Corpora

TL;DR

This paper introduces a voice conversion system that adjusts speech likability to match target preferences, using an automated predictor trained on existing data, enabling scalable and controllable voice likability modification.

Contribution

It presents a novel method for controlling voice likability in speech conversion by leveraging an automated predictor trained on large datasets, maintaining speaker identity and content.

Findings

High correlation between predictor outputs and human ratings

Effective control of voice likability demonstrated in evaluations

Preservation of speaker identity and content confirmed

Abstract

Perceived voice likability plays a crucial role in various social interactions, such as partner selection and advertising. A system that provides reference likable voice samples tailored to target audiences would enable users to adjust their speaking style and voice quality, facilitating smoother communication. To this end, we propose a voice conversion method that controls the likability of input speech while preserving both speaker identity and linguistic content. To improve training data scalability, we train a likability predictor on an existing voice likability dataset and employ it to automatically annotate a large speech synthesis corpus with likability ratings. Experimental evaluations reveal a significant correlation between the predictor's outputs and human-provided likability ratings. Subjective and objective evaluations further demonstrate that the proposed approach effectively controls voice likability while preserving both speaker identity and linguistic content.