English

ClonEval: An Open Voice Cloning Benchmark

Computation and Language 2025-09-18 v3

Abstract

We present a novel benchmark for voice cloning text-to-speech models. The benchmark consists of an evaluation protocol, an open-source library for assessing the performance of voice cloning models, and an accompanying leaderboard. The paper discusses design considerations and presents a detailed description of the evaluation procedure. The usage of the software library is explained, along with the organization of results on the leaderboard.

Cite

@article{arxiv.2504.20581,
  title  = {ClonEval: An Open Voice Cloning Benchmark},
  author = {Iwona Christop and Tomasz Kuczyński and Marek Kubis},
  journal= {arXiv preprint arXiv:2504.20581},
  year   = {2025}
}

Comments

Under review at ICASSP

R2 v1 2026-06-28T23:15:03.257Z