arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2605.25504eess.AS
收藏

面向自然情感文本到语音系统:细粒度非语言表达控制

Toward Natural Emotional Text-To-Speech System with Fine-Grained Non-Verbal Expression Control

Nara Institute of Science and Technology(奈良先端科学技术大学院大学)

Wangzixi Zhou, Bagus Tris Atmaja, Sakriani Sakti

AI总结:

提出一种基于EARS语料库的细粒度非语言表达合成方法,通过新标注方案编码NV类型、频率和时长,在情感TTS中显著提升表现力(eMOS 4.20)和情感识别准确率(78.8%)。

摘要:

虽然当前的情感文本到语音(TTS)模型已成功控制言语韵律,但它们常常忽略非语言发声(NV),而NV对于真实的人类情感至关重要。尽管最近出现了一些非语言数据集,但它们往往缺乏高质量、细粒度的标注,这限制了模型精确控制NV生成的能力。为解决这一局限,我们提出了一种细粒度非语言表达合成的新方法。我们从EARS语料库中整理并重新处理女性NV发声,开发了一种使用标签编码NV类型、频率和时长的新标注方案,并构建了一个情感TTS基准以证明其有效性。我们的评估表明,虽然我们的NV方法在感知自然度上带来轻微折衷,但它显著提升了表现力(eMOS 4.20)和情感识别准确率(78.8%)。情感特异性分析进一步揭示,NV线索对于高唤醒度情感如快乐(82.5%)和恐惧(82.7%)非常有效,并且几乎完美地传达悲伤(98.3%)。

英文摘要:

While current emotional Text-to-Speech (TTS) models have successfully controlled verbal prosody, they often ignore non-verbal vocalizations (NVs), which are essential for authentic human emotion. Although some non-verbal datasets have recently emerged, they often lack high-quality, fine-grained annotations, which restricts a model's ability to precisely control NV generation. To address this limitation, we propose a novel approach for fine-grained non-verbal expression synthesis. We curate and reprocess female NV utterances from the EARS corpus, develop a new annotation scheme using tags to encode NV types, frequencies, and durations, and build an emotional TTS benchmark to demonstrate its effectiveness. Our evaluation shows that while our NV approach leads to minor trade-offs in perceived naturalness, it significantly improves expressiveness (eMOS 4.20) and emotional recognition accuracy (78.8%). Emotion-specific analysis further reveals that NV cues are highly effective for high-arousal emotions like happy (82.5%) and fear (82.7%), and almost perfectly convey sadness (98.3%).

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。

↑