ICTNLP/WenetSpeech4TTS_manifest

Name: ICTNLP/WenetSpeech4TTS_manifest
Creator: ICTNLP
Published: 2025-05-31T03:10:44+08:00

Hugging Face2025-04-23 更新2025-05-31 收录

中文语音合成

TTS模型训练

数据链接：

https://hf-mirror.com/datasets/ICTNLP/WenetSpeech4TTS_manifest 数据链接链接失效反馈

官方服务：

资源简介：

--- {} --- Please refer to https://huggingface.co/datasets/Wenetspeech4TTS/WenetSpeech4TTS for audio files.

提供机构：

奶龙语音克隆数据集专为奶龙（NaiLong）的音色克隆和GPT-SoVITS等TTS模型训练而设计。数据集按处理阶段分为四部分：1. `raw_audio`（原始采样），包含未经处理的原始音频素材，格式为44.1kHz, 16-bit, Stereo；2. `vocal_only`（人声分离），使用UVR5的MDX-NET模型从原始音频中剥离背景音乐和环境噪音，提取干净的人声轨道；3. `slic

Hugging Face2026-04-06 更新490

voices365/Chinese_Male_017VoiceArtist_11Hours_High_Quality_Voice_Dataset

中文语音合成

情感语音生成

--- license: apache-2.0 task_categories: - text-to-speech - text-to-audio language: - zh size_categories: - 10B<n<100B --- ### Dataset Description This Dataset was recorded by a professional Chinese v

Hugging Face2024-11-04 更新110

voices365/Chinese_Male_049VoiceArtist_2Hours_High_Quality_Voice_Dataset

中文语音合成

TTS（文本转语音）

--- license: cc-by-nc-2.0 task_categories: - text-to-speech - text-to-audio - text-to-video language: - zh size_categories: - 10B<n<100B --- ### Dataset Description This dataset was recorded by a prof

Hugging Face2024-11-13 更新80

voices365/Chinese_Female_018VoiceArtist_8Hours_High_Quality_Voice_Dataset

中文语音合成

多情感语音生成

--- license: cc-by-nc-2.0 task_categories: - text-to-speech - text-to-audio - text-to-video language: - zh size_categories: - 10B<n<100B --- ### Dataset Description This Dataset was recorded by a prof

Hugging Face2024-11-04 更新120

JerryAGENDD/chinese_speech_sft_cosy_audio

中文语音合成

语音识别

这个数据集包含了文本、答案和音频三种类型的数据。文本和答案是字符串格式，音频是音频格式。数据集分为训练集，共有197,273个示例，大小为340,933,392,23.875字节。数据集下载大小为31,175,840,057字节。该数据集使用Cosy-voice 2 0.5B模型生成音频，音频输入提示来源于TigerResearch/sft_zh和JerryAGENDD/voiceprint_li

Hugging Face2025-03-04 更新90

ICTNLP/WenetSpeech4TTS_manifest

资源简介：

相关数据集