Aniket-Tathe-08/Custom_common_voice_dataset_using_RVC

该数据集是基于Common Voice v11的自定义版本，使用了Retrieval-Based Voice Conversion (RVC)技术生成自定义语音。数据集包含训练、测试、开发和其他四个分割，语言为印地语（hi），标签为XLSR。数据集的音频来源于YouTube，经过200个epoch的训练，使用了1小时的音频剪辑。数据集的许可证为CC0-1.0，属于公共领域。

This dataset is a custom variant based on Common Voice v11, where custom speech is generated using Retrieval-Based Voice Conversion (RVC) technology. It includes four splits: training, testing, development, and others. The dataset is in Hindi (hi) with the label XLSR. The audio data is sourced from YouTube, and the training was conducted over 200 epochs using 1-hour audio clips. This dataset is licensed under CC0-1.0 and is released into the public domain.

原始信息汇总

数据集概述

基本信息

• 名称: Custom Common Voice
• 语言: 印地语 (hi)
• 许可证: CC-0 1.0
• 数据集大小: 1K<n<10K
• 源数据集: common_voice_v11

数据配置

• 默认配置:
  • 训练集: transcript/hi/train.tsv
  • 测试集: transcript/hi/test.tsv
  • 其他集: transcript/hi/other.tsv
  • 开发集: transcript/hi/dev.tsv

数据来源

• 数据由RVC(Retrieval-Based Voice Conversion)创建，使用了一个自定义的语音。
• 数据来源于YouTube，具体是YouTuber Ajay Pandey的音频片段。
• 模型经过200个epoch的训练，使用了总共1小时的音频片段。

引用信息

@inproceedings{commonvoice:2020, author = {Ardila, R. and Branson, M. and Davis, K. and Henretty, M. and Kohler, M. and Meyer, J. and Morais, R. and Saunders, L. and Tyers, F. M. and Weber, G.}, title = {Common Voice: A Massively-Multilingual Speech Corpus}, booktitle = {Proceedings of the 12th Conference on Language Resources and Evaluation (LREC 2020)}, pages = {4211--4215}, year = 2020 }