Update app.py

app.py

@@ -450,7 +450,7 @@ def get_vc(sid):
             cpt = None
         return {"visible": False, "__type__": "update"}
     person = "%s/%s" % (weight_root, sid)
-    print("Загрузка %s" % person)
+    print("loading %s" % person)
     cpt = torch.load(person, map_location="cpu")
     tgt_sr = cpt["config"][-1]
     cpt["config"][-3] = cpt["weight"]["emb_g.weight"].shape[0]  # n_spk
@@ -1582,6 +1582,7 @@ with gr.Blocks(theme=gr.themes.Base(), title='Mangio-RVC-Web 💻') as app:
                             label="Mangio-Crepe Hop Length. Более высокие числа уменьшат вероятность экстремального изменения высоты тона, но более низкие числа увеличат точность. 64-192 - хороший диапазон для экспериментов.",
                             value=120,
                             interactive=True,
+                            visible=False
                             )
                         f0method0.change(fn=whethercrepeornah, inputs=[f0method0], outputs=[crepe_hop_length])
                         filter_radius0 = gr.Slider(
@@ -1616,7 +1617,125 @@ with gr.Blocks(theme=gr.themes.Base(), title='Mangio-RVC-Web 💻') as app:
                             step=0.01,
                             interactive=True,
                             )
+                            
                         
+            with gr.Accordion("Batch Conversion",open=False):
+                with gr.Row():
+                    with gr.Column():
+                        vc_transform1 = gr.Number(
+                            label=i18n("变调(整数, 半音数量, 升八度12降八度-12)"), value=0
+                        )
+                        opt_input = gr.Textbox(label=i18n("指定输出文件夹"), value="opt")
+                        f0method1 = gr.Radio(
+                            label=i18n(
+                                "选择音高提取算法,输入歌声可用pm提速,harvest低音好但巨慢无比,crepe效果好但吃GPU"
+                            ),
+                            choices=["pm", "harvest", "crepe", "rmvpe"],
+                            value="rmvpe",
+                            interactive=True,
+                        )
+                        filter_radius1 = gr.Slider(
+                            minimum=0,
+                            maximum=7,
+                            label=i18n(">=3则使用对harvest音高识别的结果使用中值滤波，数值为滤波半径，使用可以削弱哑音"),
+                            value=3,
+                            step=1,
+                            interactive=True,
+                        )
+                    with gr.Column():
+                        file_index3 = gr.Textbox(
+                            label=i18n("特征检索库文件路径,为空则使用下拉的选择结果"),
+                            value="",
+                            interactive=True,
+                        )
+                        file_index4 = gr.Dropdown(
+                            label=i18n("自动检测index路径,下拉式选择(dropdown)"),
+                            choices=sorted(index_paths),
+                            interactive=True,
+                        )
+                        refresh_button.click(
+                            fn=lambda: change_choices()[1],
+                            inputs=[],
+                            outputs=file_index4,
+                        )
+                        # file_big_npy2 = gr.Textbox(
+                        #     label=i18n("特征文件路径"),
+                        #     value="E:\\codes\\py39\\vits_vc_gpu_train\\logs\\mi-test-1key\\total_fea.npy",
+                        #     interactive=True,
+                        # )
+                        index_rate2 = gr.Slider(
+                            minimum=0,
+                            maximum=1,
+                            label=i18n("检索特征占比"),
+                            value=1,
+                            interactive=True,
+                        )
+                    with gr.Column():
+                        resample_sr1 = gr.Slider(
+                            minimum=0,
+                            maximum=48000,
+                            label=i18n("后处理重采样至最终采样率，0为不进行重采样"),
+                            value=0,
+                            step=1,
+                            interactive=True,
+                        )
+                        rms_mix_rate1 = gr.Slider(
+                            minimum=0,
+                            maximum=1,
+                            label=i18n("输入源音量包络替换输出音量包络融合比例，越靠近1越使用输出包络"),
+                            value=1,
+                            interactive=True,
+                        )
+                        protect1 = gr.Slider(
+                            minimum=0,
+                            maximum=0.5,
+                            label=i18n(
+                                "保护清辅音和呼吸声，防止电音撕裂等artifact，拉满0.5不开启，调低加大保护力度但可能降低索引效果"
+                            ),
+                            value=0.33,
+                            step=0.01,
+                            interactive=True,
+                        )
+                    with gr.Column():
+                        dir_input = gr.Textbox(
+                            label=i18n("输入待处理音频文件夹路径(去文件管理器地址栏拷就行了)"),
+                            value="E:\codes\py39\\test-20230416b\\todo-songs",
+                        )
+                        inputs = gr.File(
+                            file_count="multiple", label=i18n("也可批量输入音频文件, 二选一, 优先读文件夹")
+                        )
+                    with gr.Row():
+                        format1 = gr.Radio(
+                            label=i18n("导出文件格式"),
+                            choices=["wav", "flac", "mp3", "m4a"],
+                            value="flac",
+                            interactive=True,
+                        )
+                        but1 = gr.Button(i18n("转换"), variant="primary")
+                        vc_output3 = gr.Textbox(label=i18n("输出信息"))
+                    but1.click(
+                        vc_multi,
+                        [
+                            spk_item,
+                            dir_input,
+                            opt_input,
+                            inputs,
+                            vc_transform1,
+                            f0method1,
+                            file_index3,
+                            file_index4,
+                            # file_big_npy2,
+                            index_rate2,
+                            filter_radius1,
+                            resample_sr1,
+                            rms_mix_rate1,
+                            protect1,
+                            format1,
+                            crepe_hop_length,
+                        ],
+                        [vc_output3],
+                    )
+                    but1.click(fn=lambda: easy_uploader.clear())
         with gr.TabItem("Загрузить модель"):
             with gr.Row():
                 url=gr.Textbox(label="Введите URL-адрес модели:")
@@ -1626,6 +1745,253 @@ with gr.Blocks(theme=gr.themes.Base(), title='Mangio-RVC-Web 💻') as app:
             with gr.Row():
                 status_bar=gr.Textbox(label="")
                 download_button.click(fn=download_from_url, inputs=[url, model], outputs=[status_bar])
-        
+
+        def has_two_files_in_pretrained_folder():
+            pretrained_folder = "./pretrained/"
+            if not os.path.exists(pretrained_folder):
+                return False
+
+            files_in_folder = os.listdir(pretrained_folder)
+            num_files = len(files_in_folder)
+            return num_files >= 2
+
+        if has_two_files_in_pretrained_folder():    
+            print("Pretrained weights are downloaded. Training tab enabled!\n-------------------------------")       
+            with gr.TabItem("Train", visible=False):
+                with gr.Row():
+                    with gr.Column():
+                        exp_dir1 = gr.Textbox(label="Voice Name:", value="My-Voice")
+                        sr2 = gr.Radio(
+                            label=i18n("目标采样率"),
+                            choices=["40k", "48k"],
+                            value="40k",
+                            interactive=True,
+                            visible=False
+                        )
+                        if_f0_3 = gr.Radio(
+                            label=i18n("模型是否带音高指导(唱歌一定要, 语音可以不要)"),
+                            choices=[True, False],
+                            value=True,
+                            interactive=True,
+                            visible=False
+                        )
+                        version19 = gr.Radio(
+                            label="RVC version",
+                            choices=["v1", "v2"],
+                            value="v2",
+                            interactive=True,
+                            visible=False,
+                        )
+                        np7 = gr.Slider(
+                            minimum=0,
+                            maximum=config.n_cpu,
+                            step=1,
+                            label="# of CPUs for data processing (Leave as it is)",
+                            value=config.n_cpu,
+                            interactive=True,
+                            visible=True
+                        )
+                        trainset_dir4 = gr.Textbox(label="Path to your dataset (audios, not zip):", value="./dataset")
+                        easy_uploader = gr.Files(label='OR Drop your audios here. They will be uploaded in your dataset path above.',file_types=['audio'])
+                        but1 = gr.Button("1. Process The Dataset", variant="primary")
+                        info1 = gr.Textbox(label="Status (wait until it says 'end preprocess'):", value="")
+                        easy_uploader.upload(fn=upload_to_dataset, inputs=[easy_uploader, trainset_dir4], outputs=[info1])
+                        but1.click(
+                            preprocess_dataset, [trainset_dir4, exp_dir1, sr2, np7], [info1]
+                        )
+                    with gr.Column():
+                        spk_id5 = gr.Slider(
+                            minimum=0,
+                            maximum=4,
+                            step=1,
+                            label=i18n("请指定说话人id"),
+                            value=0,
+                            interactive=True,
+                            visible=False
+                        )
+                        with gr.Accordion('GPU Settings', open=False, visible=False):
+                            gpus6 = gr.Textbox(
+                                label=i18n("以-分隔输入使用的卡号, 例如   0-1-2   使用卡0和卡1和卡2"),
+                                value=gpus,
+                                interactive=True,
+                                visible=False
+                            )
+                            gpu_info9 = gr.Textbox(label=i18n("显卡信息"), value=gpu_info)
+                        f0method8 = gr.Radio(
+                            label=i18n(
+                                "选择音高提取算法:输入歌声可用pm提速,高质量语音但CPU差可用dio提速,harvest质量更好但慢"
+                            ),
+                            choices=["harvest","crepe", "mangio-crepe", "rmvpe"], # Fork feature: Crepe on f0 extraction for training.
+                            value="rmvpe",
+                            interactive=True,
+                        )
+                        
+                        extraction_crepe_hop_length = gr.Slider(
+                            minimum=1,
+                            maximum=512,
+                            step=1,
+                            label=i18n("crepe_hop_length"),
+                            value=128,
+                            interactive=True,
+                            visible=False,
+                        )
+                        f0method8.change(fn=whethercrepeornah, inputs=[f0method8], outputs=[extraction_crepe_hop_length])
+                        but2 = gr.Button("2. Pitch Extraction", variant="primary")
+                        info2 = gr.Textbox(label="Status(Check the Colab Notebook's cell output):", value="", max_lines=8)
+                        but2.click(
+                                extract_f0_feature,
+                                [gpus6, np7, f0method8, if_f0_3, exp_dir1, version19, extraction_crepe_hop_length],
+                                [info2],
+                            )
+                    with gr.Row():      
+                        with gr.Column():
+                            total_epoch11 = gr.Slider(
+                                minimum=1,
+                                maximum=5000,
+                                step=10,
+                                label="Total # of training epochs (IF you choose a value too high, your model will sound horribly overtrained.):",
+                                value=250,
+                                interactive=True,
+                            )
+                            butstop = gr.Button(
+                                "Stop Training",
+                                variant='primary',
+                                visible=False,
+                            )
+                            but3 = gr.Button("3. Train Model", variant="primary", visible=True)
+                            
+                            but3.click(fn=stoptraining, inputs=[gr.Number(value=0, visible=False)], outputs=[but3, butstop])
+                            butstop.click(fn=stoptraining, inputs=[gr.Number(value=1, visible=False)], outputs=[butstop, but3])
+                            
+                            
+                            but4 = gr.Button("4.Train Index", variant="primary")
+                            info3 = gr.Textbox(label="Status(Check the Colab Notebook's cell output):", value="", max_lines=10)
+                            with gr.Accordion("Training Preferences (You can leave these as they are)", open=False):
+                                #gr.Markdown(value=i18n("step3: 填写训练设置, 开始训练模型和索引"))
+                                with gr.Column():
+                                    save_epoch10 = gr.Slider(
+                                        minimum=1,
+                                        maximum=200,
+                                        step=1,
+                                        label="Backup every X amount of epochs:",
+                                        value=10,
+                                        interactive=True,
+                                    )
+                                    batch_size12 = gr.Slider(
+                                        minimum=1,
+                                        maximum=40,
+                                        step=1,
+                                        label="Batch Size (LEAVE IT unless you know what you're doing!):",
+                                        value=default_batch_size,
+                                        interactive=True,
+                                    )
+                                    if_save_latest13 = gr.Checkbox(
+                                        label="Save only the latest '.ckpt' file to save disk space.",
+                                        value=True,
+                                        interactive=True,
+                                    )
+                                    if_cache_gpu17 = gr.Checkbox(
+                                        label="Cache all training sets to GPU memory. Caching small datasets (less than 10 minutes) can speed up training, but caching large datasets will consume a lot of GPU memory and may not provide much speed improvement.",
+                                        value=False,
+                                        interactive=True,
+                                    )
+                                    if_save_every_weights18 = gr.Checkbox(
+                                        label="Save a small final model to the 'weights' folder at each save point.",
+                                        value=True,
+                                        interactive=True,
+                                    )
+                            zip_model = gr.Button('5. Download Model')
+                            zipped_model = gr.Files(label='Your Model and Index file can be downloaded here:')
+                            zip_model.click(fn=zip_downloader, inputs=[exp_dir1], outputs=[zipped_model, info3])
+                with gr.Group():
+                    with gr.Accordion("Base Model Locations:", open=False, visible=False):
+                        pretrained_G14 = gr.Textbox(
+                            label=i18n("加载预训练底模G路径"),
+                            value="pretrained_v2/f0G40k.pth",
+                            interactive=True,
+                        )
+                        pretrained_D15 = gr.Textbox(
+                            label=i18n("加载预训练底模D路径"),
+                            value="pretrained_v2/f0D40k.pth",
+                            interactive=True,
+                        )
+                        gpus16 = gr.Textbox(
+                            label=i18n("以-分隔输入使用的卡号, 例如   0-1-2   使用卡0和卡1和卡2"),
+                            value=gpus,
+                            interactive=True,
+                        )
+                    sr2.change(
+                        change_sr2,
+                        [sr2, if_f0_3, version19],
+                        [pretrained_G14, pretrained_D15, version19],
+                    )
+                    version19.change(
+                        change_version19,
+                        [sr2, if_f0_3, version19],
+                        [pretrained_G14, pretrained_D15],
+                    )
+                    if_f0_3.change(
+                        change_f0,
+                        [if_f0_3, sr2, version19],
+                        [f0method8, pretrained_G14, pretrained_D15],
+                    )
+                    but5 = gr.Button(i18n("一键训练"), variant="primary", visible=False)
+                    but3.click(
+                        click_train,
+                        [
+                            exp_dir1,
+                            sr2,
+                            if_f0_3,
+                            spk_id5,
+                            save_epoch10,
+                            total_epoch11,
+                            batch_size12,
+                            if_save_latest13,
+                            pretrained_G14,
+                            pretrained_D15,
+                            gpus16,
+                            if_cache_gpu17,
+                            if_save_every_weights18,
+                            version19,
+                        ],
+                        [
+                            info3,
+                            butstop,
+                            but3,
+                        ],
+                    )
+                    but4.click(train_index, [exp_dir1, version19], info3)
+                    but5.click(
+                        train1key,
+                        [
+                            exp_dir1,
+                            sr2,
+                            if_f0_3,
+                            trainset_dir4,
+                            spk_id5,
+                            np7,
+                            f0method8,
+                            save_epoch10,
+                            total_epoch11,
+                            batch_size12,
+                            if_save_latest13,
+                            pretrained_G14,
+                            pretrained_D15,
+                            gpus16,
+                            if_cache_gpu17,
+                            if_save_every_weights18,
+                            version19,
+                            extraction_crepe_hop_length
+                        ],
+                        info3,
+                    )
+
+        else:
+            print(
+                "Pretrained weights not downloaded. Disabling training tab.\n"
+                "Wondering how to train a voice? Visit here for the RVC model training guide: https://t.ly/RVC_Training_Guide\n"
+                "-------------------------------\n"
+            )
+                
     app.queue(concurrency_count=511, max_size=1022).launch(share=False, quiet=True)
 #endregion