Thorsten-Voice/TV-24kHz-Neutral-tokenised

该数据集是一个用于训练和微调Orpheus TTS模型家族的德语文本到语音标记化数据集。它基于大约12,000个来自原始Thorsten-Voice数据集（2022.10）的语音记录，并已重新采样到24 kHz，并使用Orpheus TTS预处理进行了标记化。数据集适用于训练和微调基于Orpheus的德语TTS模型、神经语音合成研究以及开放、无限制的TTS实验。数据集的语言是德语，说话者是Thorsten（单说话者）。音频处理细节包括重新采样到24,000 Hz，响度归一化到-24dB，并使用Orpheus TTS标记器进行标记化。数据集采用Creative Commons Zero（CC0 1.0）许可证发布，允许无限制的使用、修改、分发和构建。

This dataset is a tokenised German text-to-speech dataset created for training and fine-tuning the Orpheus TTS model family. It is based on approximately 12,000 speech recordings from the original Thorsten-Voice Dataset (2022.10) and has been resampled to 24 kHz and tokenised using Orpheus TTS preprocessing. This dataset is intended for training and fine-tuning Orpheus-based German TTS models, research on neural speech synthesis, and open, unrestricted TTS experimentation. The language of the dataset is German, and the speaker is Thorsten (single speaker). Processing details include audio resampled to 24,000 Hz, loudness normalized to -24dB, and tokenised using the Orpheus TTS tokenizer. The dataset is released under the Creative Commons Zero (CC0 1.0) license, allowing unrestricted use, modification, distribution, and building upon the dataset.