SVSNet+：利用来自语音基础模型的表示增强说话人语音相似度评估模型
SVSNet+: Enhancing Speaker Voice Similarity Assessment Models with Representations from Speech Foundation Models
AI总结：
本文提出 SVSNet+，通过融合 WavLM 等语音基础模型表示提升说话人语音相似度评估，并在 VCC 2018/2020 上验证其显著性能与跨 SFM 泛化能力。
摘要：
来自预训练语音基础模型（SFM）的表示已在许多下游任务中展现出令人瞩目的性能。然而，将预训练 SFM 表示纳入说话人语音相似度评估所能带来的潜在收益尚未得到充分研究。本文提出 SVSNet+，这是一个整合预训练 SFM 表示以提升说话人语音相似度评估性能的模型。在 Voice Conversion Challenge 2018 和 2020 数据集上的实验结果表明，纳入 WavLM 表示的 SVSNet+ 相比基线模型取得了显著改进。此外，尽管使用下游任务的小型数据集微调 WavLM 并未提升性能，但使用同一数据集学习 WavLM 的加权和表示却能够大幅提升性能。此外，当 WavLM 被其他 SFM 替换时，SVSNet+ 仍优于基线模型，并表现出很强的泛化能力。
英文摘要：
Representations from pre-trained speech foundation models (SFMs) have shown impressive performance in many downstream tasks. However, the potential benefits of incorporating pre-trained SFM representations into speaker voice similarity assessment have not been thoroughly investigated. In this paper, we propose SVSNet+, a model that integrates pre-trained SFM representations to improve performance in assessing speaker voice similarity. Experimental results on the Voice Conversion Challenge 2018 and 2020 datasets show that SVSNet+ incorporating WavLM representations shows significant improvements compared to baseline models. In addition, while fine-tuning WavLM with a small dataset of the downstream task does not improve performance, using the same dataset to learn a weighted-sum representation of WavLM can substantially improve performance. Furthermore, when WavLM is replaced by other SFMs, SVSNet+ still outperforms the baseline models and exhibits strong generalization ability.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
