WIP:DDSP-SVC

client/demo/dist/assets/gui_settings/DDSP-SVC.json

@@ -0,0 +1,186 @@
+{
+    "type": "demo",
+    "id": "DDSP-SVC",
+    "front": {
+        "title": [
+            {
+                "name": "title",
+                "options": {
+                    "mainTitle": "Realtime Voice Changer Client",
+                    "subTitle": "for DDSP-SVC",
+                    "lineNum": 1
+                }
+            },
+            {
+                "name": "clearSetting",
+                "options": {}
+            }
+        ],
+        "serverControl": [
+            {
+                "name": "startButton",
+                "options": {}
+            },
+            {
+                "name": "performance",
+                "options": {}
+            },
+            {
+                "name": "serverInfo",
+                "options": {}
+            }
+        ],
+        "modelSetting": [
+            {
+                "name": "modelUploader",
+                "options": {
+                    "showConfig": true,
+                    "showOnnx": false,
+                    "showPyTorch": true,
+                    "showCorrespondence": false,
+                    "showPyTorchCluster": false,
+
+                    "showFeature": false,
+                    "showIndex": false,
+                    "showHalfPrecision": false,
+                    "defaultEnablePyTorch": true,
+
+                    "showOnnxExportButton": false
+                }
+            },
+            {
+                "name": "framework",
+                "options": {
+                    "showFramework": true
+                }
+            },
+            {
+                "name": "modelSamplingRate",
+                "options": {}
+            }
+        ],
+        "deviceSetting": [
+            {
+                "name": "audioInput",
+                "options": {}
+            },
+            {
+                "name": "audioOutput",
+                "options": {}
+            }
+        ],
+        "qualityControl": [
+            {
+                "name": "noiseControl",
+                "options": {}
+            },
+            {
+                "name": "gainControl",
+                "options": {}
+            },
+            {
+                "name": "f0Detector",
+                "options": {
+                    "detectors": ["dio", "parselmouth", "harvest", "crepe"]
+                }
+            },
+            {
+                "name": "divider",
+                "options": {}
+            },
+            {
+                "name": "analyzer",
+                "options": {}
+            }
+        ],
+        "speakerSetting": [
+            {
+                "name": "dstId",
+                "options": {
+                    "showF0": true,
+                    "useServerInfo": false
+                }
+            },
+            {
+                "name": "tune",
+                "options": {}
+            },
+            {
+                "name": "indexRatio",
+                "options": {}
+            },
+            {
+                "name": "silentThreshold",
+                "options": {}
+            },
+            {
+                "name": "enableEnhancer",
+                "options": {}
+            }
+        ],
+        "converterSetting": [
+            {
+                "name": "inputChunkNum",
+                "options": {}
+            },
+            {
+                "name": "extraDataLength",
+                "options": {}
+            },
+            {
+                "name": "gpu",
+                "options": {}
+            }
+        ],
+        "advancedSetting": [
+            {
+                "name": "serverURL",
+                "options": {}
+            },
+            {
+                "name": "protocol",
+                "options": {}
+            },
+            {
+                "name": "sampleRate",
+                "options": {}
+            },
+            {
+                "name": "sendingSampleRate",
+                "options": {}
+            },
+            {
+                "name": "crossFadeOverlapSize",
+                "options": {}
+            },
+            {
+                "name": "crossFadeOffsetRate",
+                "options": {}
+            },
+            {
+                "name": "crossFadeEndRate",
+                "options": {}
+            },
+            {
+                "name": "downSamplingMode",
+                "options": {}
+            },
+            {
+                "name": "trancateNumThreshold",
+                "options": {}
+            }
+        ]
+    },
+
+    "dialogs": {
+        "license": [
+            {
+                "title": "Retrieval-based-Voice-Conversion-WebUI",
+                "auther": "liujing04",
+                "contact": "",
+                "url": "https://github.com/liujing04/Retrieval-based-Voice-Conversion-WebUI",
+                "license": "MIT"
+            }
+        ]
+    }
+}

client/demo/dist/assets/gui_settings/RVC.json

@@ -80,7 +80,9 @@
             },
             {
                 "name": "f0Detector",
-                "options": {}
+                "options": {
+                    "detectors": ["pm", "harvest"]
+                }
             },
             {
                 "name": "divider",

client/demo/public/assets/gui_settings/DDSP-SVC.json

@@ -0,0 +1,186 @@
+{
+    "type": "demo",
+    "id": "DDSP-SVC",
+    "front": {
+        "title": [
+            {
+                "name": "title",
+                "options": {
+                    "mainTitle": "Realtime Voice Changer Client",
+                    "subTitle": "for DDSP-SVC",
+                    "lineNum": 1
+                }
+            },
+            {
+                "name": "clearSetting",
+                "options": {}
+            }
+        ],
+        "serverControl": [
+            {
+                "name": "startButton",
+                "options": {}
+            },
+            {
+                "name": "performance",
+                "options": {}
+            },
+            {
+                "name": "serverInfo",
+                "options": {}
+            }
+        ],
+        "modelSetting": [
+            {
+                "name": "modelUploader",
+                "options": {
+                    "showConfig": true,
+                    "showOnnx": false,
+                    "showPyTorch": true,
+                    "showCorrespondence": false,
+                    "showPyTorchCluster": false,
+
+                    "showFeature": false,
+                    "showIndex": false,
+                    "showHalfPrecision": false,
+                    "defaultEnablePyTorch": true,
+
+                    "showOnnxExportButton": false
+                }
+            },
+            {
+                "name": "framework",
+                "options": {
+                    "showFramework": true
+                }
+            },
+            {
+                "name": "modelSamplingRate",
+                "options": {}
+            }
+        ],
+        "deviceSetting": [
+            {
+                "name": "audioInput",
+                "options": {}
+            },
+            {
+                "name": "audioOutput",
+                "options": {}
+            }
+        ],
+        "qualityControl": [
+            {
+                "name": "noiseControl",
+                "options": {}
+            },
+            {
+                "name": "gainControl",
+                "options": {}
+            },
+            {
+                "name": "f0Detector",
+                "options": {
+                    "detectors": ["dio", "parselmouth", "harvest", "crepe"]
+                }
+            },
+            {
+                "name": "divider",
+                "options": {}
+            },
+            {
+                "name": "analyzer",
+                "options": {}
+            }
+        ],
+        "speakerSetting": [
+            {
+                "name": "dstId",
+                "options": {
+                    "showF0": true,
+                    "useServerInfo": false
+                }
+            },
+            {
+                "name": "tune",
+                "options": {}
+            },
+            {
+                "name": "indexRatio",
+                "options": {}
+            },
+            {
+                "name": "silentThreshold",
+                "options": {}
+            },
+            {
+                "name": "enableEnhancer",
+                "options": {}
+            }
+        ],
+        "converterSetting": [
+            {
+                "name": "inputChunkNum",
+                "options": {}
+            },
+            {
+                "name": "extraDataLength",
+                "options": {}
+            },
+            {
+                "name": "gpu",
+                "options": {}
+            }
+        ],
+        "advancedSetting": [
+            {
+                "name": "serverURL",
+                "options": {}
+            },
+            {
+                "name": "protocol",
+                "options": {}
+            },
+            {
+                "name": "sampleRate",
+                "options": {}
+            },
+            {
+                "name": "sendingSampleRate",
+                "options": {}
+            },
+            {
+                "name": "crossFadeOverlapSize",
+                "options": {}
+            },
+            {
+                "name": "crossFadeOffsetRate",
+                "options": {}
+            },
+            {
+                "name": "crossFadeEndRate",
+                "options": {}
+            },
+            {
+                "name": "downSamplingMode",
+                "options": {}
+            },
+            {
+                "name": "trancateNumThreshold",
+                "options": {}
+            }
+        ]
+    },
+
+    "dialogs": {
+        "license": [
+            {
+                "title": "Retrieval-based-Voice-Conversion-WebUI",
+                "auther": "liujing04",
+                "contact": "",
+                "url": "https://github.com/liujing04/Retrieval-based-Voice-Conversion-WebUI",
+                "license": "MIT"
+            }
+        ]
+    }
+}

client/demo/public/assets/gui_settings/RVC.json

@@ -80,7 +80,9 @@
             },
             {
                 "name": "f0Detector",
-                "options": {}
+                "options": {
+                    "detectors": ["pm", "harvest"]
+                }
             },
             {
                 "name": "divider",

client/demo/src/001_ClientSelector.tsx

@@ -32,6 +32,7 @@ export const ClientSelector = () => {
                 <div></div>
                 <div className="body-button-container">
                     <div className="body-button w40 bold" onClick={() => { setClientType("RVC"); setItem(INDEXEDDB_KEY_DEFAULT_MODEL_TYPE, "RVC") }}>RVC</div>
+                    <div className="body-button w40 bold" onClick={() => { setClientType("DDSP-SVC"); setItem(INDEXEDDB_KEY_DEFAULT_MODEL_TYPE, "DDSP-SVC") }}>DDSP-SVC(N/A)</div>
                 </div>
                 <div></div>
             </div>

client/demo/src/components/demo/002_ComponentGenerator.tsx

@@ -38,6 +38,7 @@ import { RVCQualityRow, RVCQualityRowProps } from "./components/810_RVCQualityRo
 import { ModelSamplingRateRow, ModelSamplingRateRowProps } from "./components/303_ModelSamplingRateRow"
 // import { OnnxExportRow, OnnxExportRowProps } from "./components/304_OnnxExportRow"
 import { SolaEnableRow, SolaEnableRowProps } from "./components/811_SolaEnableRow"
+import { EnableEnhancerRow, EnableEnhancerRowProps } from "./components/610_EnableEnhancerRow"
 
 export const catalog: { [key: string]: (props: any) => JSX.Element } = {}
 
@@ -85,6 +86,7 @@ const initialize = () => {
     addToCatalog("noiseScale", (props: NoiseScaleRowProps) => { return <NoiseScaleRow {...props} /> })
     addToCatalog("silentThreshold", (props: SilentThresholdRowProps) => { return <SilentThresholdRow {...props} /> })
     addToCatalog("indexRatio", (props: IndexRatioRowProps) => { return <IndexRatioRow {...props} /> })
+    addToCatalog("enableEnhancer", (props: EnableEnhancerRowProps) => { return <EnableEnhancerRow {...props} /> })
 
 
 

client/demo/src/components/demo/components/301-1_ConfigSelectRow.tsx

@@ -9,7 +9,7 @@ export const ConfigSelectRow = () => {
         const configFilenameText = appState.serverSetting.fileUploadSetting.configFile?.filename || appState.serverSetting.fileUploadSetting.configFile?.file?.name || ""
         const onConfigFileLoadClicked = async () => {
             const file = await fileSelector("")
-            if (file.name.endsWith(".json") == false) {
+            if (file.name.endsWith(".json") == false && file.name.endsWith(".yaml") == false) {
                 alert("モデルファイルの拡張子はjsonである必要があります。")
                 return
             }

client/demo/src/components/demo/components/503_F0DetectorRow.tsx

@@ -4,12 +4,14 @@ import { F0Detector } from "@dannadori/voice-changer-client-js";
 
 
 export type F0DetectorRowProps = {
+    detectors: string[]
 }
-export const F0DetectorRow = (_props: F0DetectorRowProps) => {
+export const F0DetectorRow = (props: F0DetectorRowProps) => {
     const appState = useAppState()
 
     const f0DetectorRow = useMemo(() => {
         const desc = { "harvest": "harvest(High Quality)", "dio": "dio/pm(Light Weight)" }
+        const detectors = props.detectors || ["dio", "harvest"]
 
         return (
             <div className="body-row split-3-7 left-padding-1 guided">
@@ -19,9 +21,9 @@ export const F0DetectorRow = (_props: F0DetectorRowProps) => {
                         appState.serverSetting.updateServerSettings({ ...appState.serverSetting.serverSetting, f0Detector: e.target.value as F0Detector })
                     }}>
                         {
-                            Object.values(F0Detector).map(x => {
+                            Object.values(detectors).map(x => {
                                 //@ts-ignore
-                                return <option key={x} value={x}>{desc[x]}</option>
+                                return <option key={x} value={x}>{x}</option>
                             })
                         }
                     </select>

client/demo/src/components/demo/components/610_EnableEnhancerRow.tsx

@@ -0,0 +1,45 @@
+import React, { useMemo } from "react"
+import { useAppState } from "../../../001_provider/001_AppStateProvider"
+
+export type EnableEnhancerRowProps = {
+}
+
+export const EnableEnhancerRow = (_props: EnableEnhancerRowProps) => {
+    const appState = useAppState()
+
+    const clusterRatioRow = useMemo(() => {
+        return (
+            <>
+                <div className="body-row split-3-7 left-padding-1 guided">
+                    <div className="body-item-title left-padding-1 ">Enhancer</div>
+                    <div className="body-input-container">
+                        <select value={appState.serverSetting.serverSetting.enableEnhancer} onChange={(e) => {
+                            appState.serverSetting.updateServerSettings({ ...appState.serverSetting.serverSetting, enableEnhancer: Number(e.target.value) })
+                        }}>
+                            <option value="0" >disable</option>
+                            <option value="1" >enable</option>
+                        </select>
+                    </div>
+                </div>
+                <div className="body-row split-3-3-4 left-padding-1 guided">
+                    <div className="body-item-title left-padding-1 ">Enhancer Tune</div>
+                    <div>
+                        <input type="range" className="body-item-input-slider" min="0" max="10" step="1" value={appState.serverSetting.serverSetting.enhancerTune || 0} onChange={(e) => {
+                            appState.serverSetting.updateServerSettings({ ...appState.serverSetting.serverSetting, enhancerTune: Number(e.target.value) })
+                        }}></input>
+                        <span className="body-item-input-slider-val">{appState.serverSetting.serverSetting.enhancerTune}</span>
+                    </div>
+                    <div className="body-button-container">
+                    </div>
+                </div>
+            </>
+        )
+    }, [
+        appState.serverSetting.serverSetting,
+        appState.serverSetting.updateServerSettings
+    ])
+
+    return clusterRatioRow
+}
+
+

client/lib/src/const.ts

@@ -10,6 +10,7 @@ export const ClientType = {
     "so-vits-svc-40": "so-vits-svc-40",
     "so-vits-svc-40_c": "so-vits-svc-40_c",
     "so-vits-svc-40v2": "so-vits-svc-40v2",
+    "DDSP-SVC": "DDSP-SVC",
     "RVC": "RVC"
 
 } as const
@@ -58,6 +59,8 @@ export type Framework = typeof Framework[keyof typeof Framework]
 export const F0Detector = {
     "dio": "dio",
     "harvest": "harvest",
+    "parselmouth": "parselmouth",
+    "crepe": "crepe",
 } as const
 export type F0Detector = typeof F0Detector[keyof typeof F0Detector]
 
@@ -91,6 +94,8 @@ export const ServerSettingKey = {
     "rvcQuality": "rvcQuality",
     "modelSamplingRate": "modelSamplingRate",
 
+    "enableEnhancer": "enableEnhancer",
+    "enhancerTune": "enhancerTune",
 
     "inputSampleRate": "inputSampleRate",
 } as const
@@ -125,6 +130,9 @@ export type VoiceChangerServerSetting = {
     rvcQuality: number // 0:low, 1:high
     modelSamplingRate: ModelSamplingRate // 32000,40000,48000
 
+    enableEnhancer: number // DDSP-SVC
+    enhancerTune: number // DDSP-SVC
+
     inputSampleRate: InputSampleRate
 }
 
@@ -166,6 +174,9 @@ export const DefaultServerSetting_MMVCv15: ServerInfo = {
     rvcQuality: 0,
     modelSamplingRate: 48000,
 
+    enableEnhancer: 0,
+    enhancerTune: 0,
+
     inputSampleRate: 24000,
 
     // 
@@ -203,6 +214,8 @@ export const DefaultServerSetting_MMVCv13: ServerInfo = {
     rvcQuality: 0,
     modelSamplingRate: 48000,
 
+    enableEnhancer: 0,
+    enhancerTune: 0,
 
     inputSampleRate: 24000,
 
@@ -245,6 +258,8 @@ export const DefaultServerSetting_so_vits_svc_40: ServerInfo = {
     rvcQuality: 0,
     modelSamplingRate: 48000,
 
+    enableEnhancer: 0,
+    enhancerTune: 0,
 
     inputSampleRate: 24000,
 
@@ -287,6 +302,8 @@ export const DefaultServerSetting_so_vits_svc_40_c: ServerInfo = {
     rvcQuality: 0,
     modelSamplingRate: 48000,
 
+    enableEnhancer: 0,
+    enhancerTune: 0,
 
     inputSampleRate: 24000,
 
@@ -328,6 +345,52 @@ export const DefaultServerSetting_so_vits_svc_40v2: ServerInfo = {
     rvcQuality: 0,
     modelSamplingRate: 48000,
 
+    enableEnhancer: 0,
+    enhancerTune: 0,
+
+    inputSampleRate: 24000,
+
+    // 
+    status: "ok",
+    configFile: "",
+    pyTorchModelFile: "",
+    onnxModelFile: "",
+    onnxExecutionProviders: []
+}
+
+export const DefaultServerSetting_DDSP_SVC: ServerInfo = {
+    srcId: 0,
+    dstId: 0,
+    gpu: 0,
+
+    crossFadeOffsetRate: 0.0,
+    crossFadeEndRate: 1.0,
+    crossFadeOverlapSize: CrossFadeOverlapSize[1024],
+    solaEnabled: 0,
+
+    framework: Framework.PyTorch,
+    f0Factor: 1.0,
+    onnxExecutionProvider: OnnxExecutionProvider.CPUExecutionProvider,
+    f0Detector: F0Detector.dio,
+    recordIO: 0,
+
+    // tran: 0,
+    // noiceScale: 0,
+    // predictF0: 0,
+    // silentThreshold: 0,
+    tran: 10,
+    noiceScale: 0.3,
+    predictF0: 0,
+    silentThreshold: 0.00001,
+    extraConvertSize: 1024 * 32,
+    clusterInferRatio: 0.1,
+
+    indexRatio: 0,
+    rvcQuality: 0,
+    modelSamplingRate: 48000,
+
+    enableEnhancer: 0,
+    enhancerTune: 0,
 
     inputSampleRate: 24000,
 
@@ -371,6 +434,9 @@ export const DefaultServerSetting_RVC: ServerInfo = {
     rvcQuality: 0,
     modelSamplingRate: 48000,
 
+    enableEnhancer: 0,
+    enhancerTune: 0,
+
     inputSampleRate: 48000,
 
     // 
@@ -449,6 +515,13 @@ export const DefaultWorkletNodeSetting_so_vits_svc_40v2: WorkletNodeSetting = {
     downSamplingMode: "average"
 }
 
+export const DefaultWorkletNodeSetting_DDSP_SVC: WorkletNodeSetting = {
+    serverUrl: "",
+    protocol: "sio",
+    sendingSampleRate: 48000,
+    inputChunkNum: 256,
+    downSamplingMode: "average"
+}
 
 export const DefaultWorkletNodeSetting_RVC: WorkletNodeSetting = {
     serverUrl: "",

client/lib/src/hooks/useServerSetting.ts

@@ -1,5 +1,5 @@
 import { useState, useMemo, useEffect } from "react"
-import { VoiceChangerServerSetting, ServerInfo, ServerSettingKey, INDEXEDDB_KEY_SERVER, INDEXEDDB_KEY_MODEL_DATA, ClientType, DefaultServerSetting_MMVCv13, DefaultServerSetting_MMVCv15, DefaultServerSetting_so_vits_svc_40v2, DefaultServerSetting_so_vits_svc_40, DefaultServerSetting_so_vits_svc_40_c, DefaultServerSetting_RVC, OnnxExporterInfo } from "../const"
+import { VoiceChangerServerSetting, ServerInfo, ServerSettingKey, INDEXEDDB_KEY_SERVER, INDEXEDDB_KEY_MODEL_DATA, ClientType, DefaultServerSetting_MMVCv13, DefaultServerSetting_MMVCv15, DefaultServerSetting_so_vits_svc_40v2, DefaultServerSetting_so_vits_svc_40, DefaultServerSetting_so_vits_svc_40_c, DefaultServerSetting_RVC, OnnxExporterInfo, DefaultServerSetting_DDSP_SVC } from "../const"
 import { VoiceChangerClient } from "../VoiceChangerClient"
 import { useIndexedDB } from "./useIndexedDB"
 
@@ -72,6 +72,8 @@ export const useServerSetting = (props: UseServerSettingProps): ServerSettingSta
             return DefaultServerSetting_so_vits_svc_40_c
         } else if (props.clientType == "so-vits-svc-40v2") {
             return DefaultServerSetting_so_vits_svc_40v2
+        } else if (props.clientType == "DDSP-SVC") {
+            return DefaultServerSetting_DDSP_SVC
         } else if (props.clientType == "RVC") {
             return DefaultServerSetting_RVC
         } else {

client/lib/src/hooks/useWorkletNodeSetting.ts

@@ -1,6 +1,6 @@
 import { useState, useMemo, useEffect } from "react"
 
-import { ClientType, DefaultWorkletNodeSetting, DefaultWorkletNodeSetting_RVC, DefaultWorkletNodeSetting_so_vits_svc_40, DefaultWorkletNodeSetting_so_vits_svc_40v2, INDEXEDDB_KEY_WORKLETNODE, WorkletNodeSetting } from "../const"
+import { ClientType, DefaultWorkletNodeSetting, DefaultWorkletNodeSetting_DDSP_SVC, DefaultWorkletNodeSetting_RVC, DefaultWorkletNodeSetting_so_vits_svc_40, DefaultWorkletNodeSetting_so_vits_svc_40v2, INDEXEDDB_KEY_WORKLETNODE, WorkletNodeSetting } from "../const"
 import { VoiceChangerClient } from "../VoiceChangerClient"
 import { useIndexedDB } from "./useIndexedDB"
 
@@ -28,6 +28,8 @@ export const useWorkletNodeSetting = (props: UseWorkletNodeSettingProps): Workle
             return DefaultWorkletNodeSetting_so_vits_svc_40
         } else if (props.clientType == "so-vits-svc-40v2") {
             return DefaultWorkletNodeSetting_so_vits_svc_40v2
+        } else if (props.clientType == "DDSP-SVC") {
+            return DefaultWorkletNodeSetting_DDSP_SVC
         } else if (props.clientType == "RVC") {
             return DefaultWorkletNodeSetting_RVC
         } else {

server/const.py

@@ -40,17 +40,4 @@ def setModelType(_modelType: ModelType):
 
 def getFrontendPath():
     frontend_path = os.path.join(sys._MEIPASS, "dist") if hasattr(sys, "_MEIPASS") else "../client/demo/dist"
-    # if modelType == "MMVCv15":
-
-    # elif modelType == "MMVCv13":
-    #     frontend_path = os.path.join(sys._MEIPASS, "dist_v13") if hasattr(sys, "_MEIPASS") else "../client/demo_v13/dist"
-    # elif modelType == "so-vits-svc-40":
-    #     frontend_path = os.path.join(sys._MEIPASS, "dist_v13") if hasattr(sys, "_MEIPASS") else "../client/demo_v13/dist"
-    # elif modelType == "so-vits-svc-40v2":
-    #     frontend_path = os.path.join(sys._MEIPASS, "dist_v13") if hasattr(sys, "_MEIPASS") else "../client/demo_v13/dist"
-    # elif modelType == "DDSP-SVC":
-    #     frontend_path = os.path.join(sys._MEIPASS, "dist_v13") if hasattr(sys, "_MEIPASS") else "../client/demo_v13/dist"
-    # elif modelType == "so-vits-svc-40_c":
-    #     frontend_path = os.path.join(sys._MEIPASS, "dist_v13") if hasattr(sys, "_MEIPASS") \
-    #         else "../client/demo_v13/dist"
     return frontend_path

server/voice_changer/DDSP_SVC/DDSP_SVC.py

@@ -34,13 +34,14 @@ class DDSP_SVCSettings():
     gpu: int = 0
     dstId: int = 0
 
-    f0Detector: str = "dio"  # dio or harvest
+    f0Detector: str = "dio"  # dio or harvest # parselmouth
     tran: int = 20
-    noiceScale: float = 0.3
     predictF0: int = 0  # 0:False, 1:True
     silentThreshold: float = 0.00001
     extraConvertSize: int = 1024 * 32
-    clusterInferRatio: float = 0.1
+
+    enableEnhancer: int = 0
+    enhancerTune: int = 0
 
     framework: str = "PyTorch"  # PyTorch or ONNX
     pyTorchModelFile: str = ""
@@ -52,7 +53,7 @@ class DDSP_SVCSettings():
     )
 
     # ↓mutableな物だけ列挙
-    intData = ["gpu", "dstId", "tran", "predictF0", "extraConvertSize"]
+    intData = ["gpu", "dstId", "tran", "predictF0", "extraConvertSize", "enableEnhancer", "enhancerTune"]
     floatData = ["noiceScale", "silentThreshold", "clusterInferRatio"]
     strData = ["framework", "f0Detector"]
 
@@ -63,23 +64,24 @@ class DDSP_SVC:
         self.net_g = None
         self.onnx_session = None
 
-        self.raw_path = io.BytesIO()
         self.gpu_num = torch.cuda.device_count()
         self.prevVol = 0
         self.params = params
         print("DDSP-SVC initialization:", params)
 
-    def loadModel(self, config: str, pyTorch_model_file: str = None, onnx_model_file: str = None, clusterTorchModel: str = None):
+    def loadModel(self, props):
-
+        self.settings.configFile = props["files"]["configFilename"]
-        self.settings.configFile = config
+        self.settings.pyTorchModelFile = props["files"]["pyTorchModelFilename"]
         # model
-        model, args = vo.load_model(pyTorch_model_file)
+        model, args = vo.load_model(self.settings.pyTorchModelFile)
         self.model = model
         self.args = args
         self.hop_size = int(self.args.data.block_size * SAMPLING_RATE / self.args.data.sampling_rate)
+        # self.sampling_rate = args.data.sampling_rate
 
         # hubert
-        vec_path = self.params["hubert"]
+        # vec_path = self.params["hubert"]
+        vec_path = "./model_DDSP-SVC/hubert-soft-0d54a1f4.pt"
         self.encoder = vo.Units_Encoder(
             args.data.encoder,
             vec_path,
@@ -134,6 +136,16 @@ class DDSP_SVC:
             setattr(self.settings, key, float(val))
         elif key in self.settings.strData:
             setattr(self.settings, key, str(val))
+            if key == "f0Detector":
+                print("f0Detector update", val)
+                if val == "dio":
+                    val = "parselmouth"
+                self.f0_detector = vo.F0_Extractor(
+                    val,
+                    SAMPLING_RATE,
+                    self.hop_size,
+                    float(50),
+                    float(1100))
         else:
             return False
 
@@ -155,7 +167,7 @@ class DDSP_SVC:
     def get_processing_sampling_rate(self):
         return SAMPLING_RATE
 
-    def generate_input(self, newData: any, inputSize: int, crossfadeSize: int):
+    def generate_input(self, newData: any, inputSize: int, crossfadeSize: int, solaSearchFrame: int = 0):
         newData = newData.astype(np.float32) / 32768.0
 
         if hasattr(self, "audio_buffer"):
@@ -163,7 +175,8 @@ class DDSP_SVC:
         else:
             self.audio_buffer = newData
 
-        convertSize = inputSize + crossfadeSize + self.settings.extraConvertSize
+        convertSize = inputSize + crossfadeSize + solaSearchFrame + self.settings.extraConvertSize
+
         if convertSize % self.hop_size != 0:  # モデルの出力のホップサイズで切り捨てが発生するので補う。
             convertSize = convertSize + (self.hop_size - (convertSize % self.hop_size))
 
@@ -228,8 +241,6 @@ class DDSP_SVC:
 
         return result
 
-        pass
-
     def _pyTorch_inference(self, data):
 
         if hasattr(self, "model") == False or self.model == None:
@@ -244,6 +255,8 @@ class DDSP_SVC:
         convertSize = data[4]
         vol = data[5]
 
+        print(volume.device)
+
         # if vol < self.settings.silentThreshold:
         #     print("threshold")
         #     return np.zeros(convertSize).astype(np.int16)
@@ -253,12 +266,14 @@ class DDSP_SVC:
             seg_output, _, (s_h, s_n) = self.model(c, f0, volume, spk_id=spk_id, spk_mix_dict=None)
             seg_output *= mask
 
-            seg_output, output_sample_rate = self.enhancer.enhance(
+            if self.settings.enableEnhancer:
-                seg_output,
+                seg_output, output_sample_rate = self.enhancer.enhance(
-                self.args.data.sampling_rate,
+                    seg_output,
-                f0,
+                    self.args.data.sampling_rate,
-                self.args.data.block_size,
+                    f0,
-                adaptive_key=float(3))
+                    self.args.data.block_size,
+                    adaptive_key=float(self.settings.enhancerTune))
+
             result = seg_output.squeeze().cpu().numpy() * 32768.0
         return np.array(result).astype(np.int16)
 
@@ -277,12 +292,15 @@ class DDSP_SVC:
         del self.net_g
         del self.onnx_session
 
+        remove_path = os.path.join("DDSP-SVC")
+        sys.path = [x for x in sys.path if x.endswith(remove_path) == False]
 
-def cross_fade(a: np.ndarray, b: np.ndarray, idx: int):
+        for key in list(sys.modules):
-    result = np.zeros(idx + b.shape[0])
+            val = sys.modules.get(key)
-    fade_len = a.shape[0] - idx
+            try:
-    np.copyto(dst=result[:idx], src=a[:idx])
+                file_path = val.__file__
-    k = np.linspace(0, 1.0, num=fade_len, endpoint=True)
+                if file_path.find("DDSP-SVC" + os.path.sep) >= 0:
-    result[idx: a.shape[0]] = (1 - k) * a[idx:] + k * b[: fade_len]
+                    print("remove", key, file_path)
-    np.copyto(dst=result[a.shape[0]:], src=b[fade_len:])
+                    sys.modules.pop(key)
-    return result
+            except Exception as e:
+                pass