Update README.md
Browse files
README.md
CHANGED
@@ -25,15 +25,15 @@ tags:
|
|
25 |
|
26 |
取自岁己22年12月、23年1月的录播(除电台,共计211:13:21),经过以下步骤筛选处理
|
27 |
|
28 |
-
1. 挑取BGM音量较低的片段(20:39:21)
|
29 |
|
30 |
-
2. [UVR5](https://github.com/Anjok07/ultimatevocalremovergui) VR Architecture 5_HP-Karaoke-UVR处理,尽量除去了BGM中的人声(20:39:20,反正确实就是少了1s)
|
31 |
|
32 |
-
3. [Audio Slicer](https://github.com/flutydeer/audio-slicer)切片(12:45:29)
|
33 |
|
34 |
-
4. [Fish Audio Preprocessor](https://github.com/fishaudio/audio-preprocess)响度标准化并删除过短过长的片段(11:24:06)
|
35 |
|
36 |
-
5. [Spliter Wav by IceKyrin](https://github.com/IceKyrin)声纹识别稳定数据(06:47:46)
|
37 |
|
38 |
文件结构:
|
39 |
|
|
|
25 |
|
26 |
取自岁己22年12月、23年1月的录播(除电台,共计211:13:21),经过以下步骤筛选处理
|
27 |
|
28 |
+
1. 挑取BGM音量较低的片段(20:39:21)_[[LowBGM.zip]](https://huggingface.co/datasets/Miuzarte/SUISovitsDataForBaseModel/blob/main/%E6%9C%89%E7%9A%84%E6%B2%A1%E7%9A%84/LowBGM.zip)
|
29 |
|
30 |
+
2. [UVR5](https://github.com/Anjok07/ultimatevocalremovergui) VR Architecture 5_HP-Karaoke-UVR处理,尽量除去了BGM中的人声(20:39:20,反正确实就是少了1s)_[[UVR-ed.zip]](https://huggingface.co/datasets/Miuzarte/SUISovitsDataForBaseModel/blob/main/%E6%9C%89%E7%9A%84%E6%B2%A1%E7%9A%84/UVR-ed.zip)
|
31 |
|
32 |
+
3. [Audio Slicer](https://github.com/flutydeer/audio-slicer)切片(12:45:29)_[[Slice-d.zip]](https://huggingface.co/datasets/Miuzarte/SUISovitsDataForBaseModel/blob/main/%E6%9C%89%E7%9A%84%E6%B2%A1%E7%9A%84/Slice-d.zip)
|
33 |
|
34 |
+
4. [Fish Audio Preprocessor](https://github.com/fishaudio/audio-preprocess)响度标准化并删除过短过长的片段(11:24:06)_[[LoudnessNorm-ed.zip]](https://huggingface.co/datasets/Miuzarte/SUISovitsDataForBaseModel/blob/main/%E6%9C%89%E7%9A%84%E6%B2%A1%E7%9A%84/LoudnessNorm-ed.zip)
|
35 |
|
36 |
+
5. [Spliter Wav by IceKyrin](https://github.com/IceKyrin)声纹识别稳定数据(06:47:46)_[[ForBaseModel.zip]](https://huggingface.co/datasets/Miuzarte/SUISovitsDataForBaseModel/blob/main/ForBaseModel.zip)
|
37 |
|
38 |
文件结构:
|
39 |
|