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 2501.15907cs.SDcs.CLeess.AS
收藏

Emilia:用于语音生成的大规模、广泛、多语言和多样化数据集

Emilia: A Large-Scale, Extensive, Multilingual, and Diverse Dataset for Speech Generation

Chinese University of Hong Kong, Shenzhen(香港中文大学(深圳));Laboratory of Speech and Intelligent Information Processing, Institute of Acoustics, CAS(声学研究所语音与智能信息处理实验室);University of Chinese Academy of Sciences(中国科学院大学);Shanghai AI Laboratory(上海人工智能实验室)

Haorui He, Zengqiang Shang, Chaoren Wang, Xuyuan Li, Yicheng Gu, Hua Hua, Liwei Liu, Chen Yang, Jiaqi Li, Peiyang Shi, Yuancheng Wang, Kai Chen, Pengyuan Zhang, Zhizheng Wu

AI总结:

本文提出Emilia-Pipe流水线构建大规模多语言语音数据集Emilia(超10.1万小时)及Emilia-Large(超21.6万小时),训练模型生成更自发、类人的语音,验证了数据规模对语音生成性能的关键作用。

摘要:

近期语音生成的进展得益于大规模训练数据集。然而,当前模型难以捕捉真实人类语音中固有的自发性和多变性,因为它们主要基于有声书数据集训练,这些数据集仅限于正式的、朗读式的说话风格。为解决这一局限,我们引入了Emilia-Pipe,一个开源预处理流水线,旨在从宝贵但尚未充分开发的野外来源中提取高质量训练数据,这些来源捕捉了真实世界情境中的自发人类语音。利用Emilia-Pipe,我们构建了Emilia数据集,包含超过10.1万小时的语音,涵盖六种语言:英语、中文、德语、法语、日语和韩语。此外,我们将Emilia扩展为Emilia-Large数据集,超过21.6万小时,使其成为可用的最大开源语音生成资源之一。大量实验表明,与在传统有声书数据集上训练的模型相比,Emilia训练的模型生成的语音明显更自发、更接近人类,同时保持了相当的可懂度。这些模型更好地捕捉了多样的说话人音色和真实世界对话风格的完整谱系。我们的工作还强调了扩大数据集规模对于提升语音生成性能的重要性,并验证了Emilia在多语言和跨语言语音生成任务中的有效性。

英文摘要:

Recent advancements in speech generation have been driven by large-scale training datasets. However, current models struggle to capture the spontaneity and variability inherent in real-world human speech, as they are primarily trained on audio-book datasets limited to formal, read-aloud speaking styles. To address this limitation, we introduce Emilia-Pipe, an open-source preprocessing pipeline designed to extract high-quality training data from valuable yet under-explored in-the-wild sources that capture spontaneous human speech in real-world contexts. Using Emilia-Pipe, we construct Emilia, which comprises over 101k hours of speech across six languages: English, Chinese, German, French, Japanese, and Korean. Furthermore, we expand Emilia to Emilia-Large, a dataset exceeding 216k hours, making it one of the largest open-source speech generation resources available. Extensive experiments show that Emilia-trained models produce markedly more spontaneous, human-like speech than those trained on traditional audio-book datasets, while matching their intelligibility. These models better capture diverse speaker timbres and the full spectrum of real-world conversational styles. Our work also highlights the importance of scaling dataset size for advancing speech generation performance and validates the effectiveness of Emilia for both multilingual and crosslingual speech generation tasks.

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。

↑