通过自净化流匹配实现鲁棒TTS训练:用于WildSpoof 2026 TTS赛道
Robust TTS Training via Self-Purifying Flow Matching for the WildSpoof 2026 TTS Track
AI总结:
本文提出通过自净化流匹配技术提升TTS系统在野外语音中的鲁棒性,使其在WildSpoof 2026 TTS赛道中取得优异成绩。
摘要:
本文提出了一种轻量级的文本到语音(TTS)系统,用于WildSpoof挑战TTS赛道。我们的方法通过自净化流匹配(SPFM)对最近发布的开源TTS模型Supertonic进行微调,以实现对野外语音的鲁棒适应。SPFM通过在每个样本上比较条件性和无条件性流匹配损失来缓解标签噪声,将可疑的文本-语音对路由到无条件训练,同时利用其语音信息。所得到的模型在所有参赛队伍中实现了最低的词错误率(WER),并在感知指标如UTMOS和DNSMOS上排名第二。这些发现表明,当结合显式的噪声处理机制如SPFM时,高效的开源架构如Supertonic可以有效适应多样的现实语音条件。
英文摘要:
This paper presents a lightweight text-to-speech (TTS) system developed for the WildSpoof Challenge TTS Track. Our approach fine-tunes the recently released open-weight TTS model, \textit{Supertonic}\footnote{\url{https://github.com/supertone-inc/supertonic}}, with Self-Purifying Flow Matching (SPFM) to enable robust adaptation to in-the-wild speech. SPFM mitigates label noise by comparing conditional and unconditional flow matching losses on each sample, routing suspicious text--speech pairs to unconditional training while still leveraging their acoustic information. The resulting model achieves the lowest Word Error Rate (WER) among all participating teams, while ranking second in perceptual metrics such as UTMOS and DNSMOS. These findings demonstrate that efficient, open-weight architectures like Supertonic can be effectively adapted to diverse real-world speech conditions when combined with explicit noise-handling mechanisms such as SPFM.
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。