Optimize user interaction for feature retrieval

README.md

@@ -338,7 +338,7 @@ Optional parameters: see the next section
 - `-lg` | `--linear_gradient`: The cross fade length of two audio slices in seconds. If there is a discontinuous voice after forced slicing, you can adjust this value. Otherwise, it is recommended to use the default value of 0.
 - `-f0p` | `--f0_predictor`: Select a F0 predictor, options are `crepe`, `pm`, `dio`, `harvest`, default value is `pm`(note: f0 mean pooling will be enable when using `crepe`)
 - `-a` | `--auto_predict_f0`: automatic pitch prediction, do not enable this when converting singing voices as it can cause serious pitch issues.
-- `-cm` | `--cluster_model_path`: Cluster model or feature retrieval index path, if there is no training cluster or feature retrieval, fill in at will.
+- `-cm` | `--cluster_model_path`: Cluster model or feature retrieval index path, if left blank, it will be automatically set as the default path of these models. If there is no training cluster or feature retrieval, fill in at will.
 - `-cr` | `--cluster_infer_ratio`: The proportion of clustering scheme or feature retrieval ranges from 0 to 1. If there is no training clustering model or feature retrieval, the default is 0.
 - `-eh` | `--enhance`: Whether to use NSF_HIFIGAN enhancer, this option has certain effect on sound quality enhancement for some models with few training sets, but has negative effect on well-trained models, so it is disabled by default.
 - `-shd` | `--shallow_diffusion`: Whether to use shallow diffusion, which can solve some electrical sound problems after use. This option is disabled by default. When this option is enabled, NSF_HIFIGAN enhancer will be disabled
@@ -378,7 +378,7 @@ No changes are required in the existing steps. Simply train an additional cluste
   - Execute `python cluster/train_cluster.py`. The output model will be saved in `logs/44k/kmeans_10000.pt`.
   - The clustering model can currently be trained using the gpu by executing `python cluster/train_cluster.py --gpu`
 - Inference process:
-  - Specify `cluster_model_path` in `inference_main.py`.
+  - Specify `cluster_model_path` in `inference_main.py`. If not specified, the default is `logs/44k/kmeans_10000.pt`.
   - Specify `cluster_infer_ratio` in `inference_main.py`, where `0` means not using clustering at all, `1` means only using clustering, and usually `0.5` is sufficient.
 
 ### Feature retrieval
@@ -396,7 +396,7 @@ The output of the model will be in `logs/44k/feature_and_index.pkl`
 
 - Inference process: 
   - The `--feature_retrieval` needs to be formulated first, and the clustering mode automatically switches to the feature retrieval mode.
-  - Specify `cluster_model_path` in `inference_main.py`.
+  - Specify `cluster_model_path` in `inference_main.py`. If not specified, the default is `logs/44k/feature_and_index.pkl`.
   - Specify `cluster_infer_ratio` in `inference_main.py`, where `0` means not using feature retrieval at all, `1` means only using feature retrieval, and usually `0.5` is sufficient.
 
 ### [![Open in Colab](https://colab.research.google.com/assets/colab-badge.svg)](https://colab.research.google.com/github/svc-develop-team/so-vits-svc/blob/4.1-Stable/sovits4_for_colab.ipynb) [sovits4_for_colab.ipynb](https://colab.research.google.com/github/svc-develop-team/so-vits-svc/blob/4.1-Stable/sovits4_for_colab.ipynb)

README_zh_CN.md

@@ -340,7 +340,7 @@ python inference_main.py -m "logs/44k/G_30400.pth" -c "configs/config.json" -n "
 + `-lg` | `--linear_gradient`：两段音频切片的交叉淡入长度，如果强制切片后出现人声不连贯可调整该数值，如果连贯建议采用默认值0，单位为秒
 + `-f0p` | `--f0_predictor`：选择F0预测器,可选择crepe,pm,dio,harvest,默认为pm(注意：crepe为原F0使用均值滤波器)
 + `-a` | `--auto_predict_f0`：语音转换自动预测音高，转换歌声时不要打开这个会严重跑调
-+ `-cm` | `--cluster_model_path`：聚类模型或特征检索索引路径，如果没有训练聚类或特征检索则随便填
++ `-cm` | `--cluster_model_path`：聚类模型或特征检索索引路径，留空则自动设为各方案模型的默认路径，如果没有训练聚类或特征检索则随便填
 + `-cr` | `--cluster_infer_ratio`：聚类方案或特征检索占比，范围0-1，若没有训练聚类模型或特征检索则默认0即可
 + `-eh` | `--enhance`：是否使用NSF_HIFIGAN增强器,该选项对部分训练集少的模型有一定的音质增强效果，但是对训练好的模型有反面效果，默认关闭
 + `-shd` | `--shallow_diffusion`：是否使用浅层扩散，使用后可解决一部分电音问题，默认关闭，该选项打开时，NSF_HIFIGAN增强器将会被禁止
@@ -379,7 +379,7 @@ python inference_main.py -m "logs/44k/G_30400.pth" -c "configs/config.json" -n "
   + 执行`python cluster/train_cluster.py`，模型的输出会在`logs/44k/kmeans_10000.pt`
   + 聚类模型目前可以使用gpu进行训练，执行`python cluster/train_cluster.py --gpu`
 + 推理过程：
-  + `inference_main.py`中指定`cluster_model_path`
+  + `inference_main.py`中指定`cluster_model_path` 为模型输出文件, 留空则默认为`logs/44k/kmeans_10000.pt`
   + `inference_main.py`中指定`cluster_infer_ratio`，`0`为完全不使用聚类，`1`为只使用聚类，通常设置`0.5`即可
 
 ### 特征检索
@@ -397,7 +397,7 @@ python train_index.py -c configs/config.json
 
 + 推理过程：
   + 需要首先制定`--feature_retrieval`，此时聚类方案会自动切换到特征检索方案
-  + `inference_main.py`中指定`cluster_model_path` 为模型输出文件
+  + `inference_main.py`中指定`cluster_model_path` 为模型输出文件, 留空则默认为`logs/44k/feature_and_index.pkl`
   + `inference_main.py`中指定`cluster_infer_ratio`，`0`为完全不使用特征检索，`1`为只使用特征检索，通常设置`0.5`即可
 
 ### [![Open in Colab](https://colab.research.google.com/assets/colab-badge.svg)](https://colab.research.google.com/github/svc-develop-team/so-vits-svc/blob/4.1-Stable/sovits4_for_colab.ipynb) [sovits4_for_colab.ipynb](https://colab.research.google.com/github/svc-develop-team/so-vits-svc/blob/4.1-Stable/sovits4_for_colab.ipynb)

inference_main.py

@@ -31,7 +31,7 @@ def main():
     
     # 可选项部分
     parser.add_argument('-a', '--auto_predict_f0', action='store_true', default=False, help='语音转换自动预测音高，转换歌声时不要打开这个会严重跑调')
-    parser.add_argument('-cm', '--cluster_model_path', type=str, default="logs/44k/kmeans_10000.pt", help='聚类模型或特征检索索引路径，如果没有训练聚类或特征检索则随便填')
+    parser.add_argument('-cm', '--cluster_model_path', type=str, default="", help='聚类模型或特征检索索引路径，留空则自动设为各方案模型的默认路径，如果没有训练聚类或特征检索则随便填')
     parser.add_argument('-cr', '--cluster_infer_ratio', type=float, default=0, help='聚类方案或特征检索占比，范围0-1，若没有训练聚类模型或特征检索则默认0即可')
     parser.add_argument('-lg', '--linear_gradient', type=float, default=0, help='两段音频切片的交叉淡入长度，如果强制切片后出现人声不连贯可调整该数值，如果连贯建议采用默认值0，单位为秒')
     parser.add_argument('-f0p', '--f0_predictor', type=str, default="pm", help='选择F0预测器,可选择crepe,pm,dio,harvest,默认为pm(注意：crepe为原F0使用均值滤波器)')
@@ -87,6 +87,15 @@ def main():
     second_encoding = args.second_encoding
     loudness_envelope_adjustment = args.loudness_envelope_adjustment
 
+    if cluster_infer_ratio != 0:
+        if args.cluster_model_path == "":
+            if args.feature_retrieval:  # 若指定了占比但没有指定模型路径，则按是否使用特征检索分配默认的模型路径
+                args.cluster_model_path = "logs/44k/feature_and_index.pkl"
+            else:
+                args.cluster_model_path = "logs/44k/kmeans_10000.pt"
+    else:  # 若未指定占比，则无论是否指定模型路径，都将其置空以避免之后的模型加载
+        args.cluster_model_path = ""
+
     svc_model = Svc(args.model_path,
                     args.config_path,
                     args.device,