mirror of
https://github.com/w-okada/voice-changer.git
synced 2025-01-23 05:25:01 +03:00
update
This commit is contained in:
w-okada
parent
bc6e8a9c08
commit
2ac5ec9feb
11
client/.vscode/settings.json
vendored
Normal file
11
client/.vscode/settings.json
vendored
Normal file
@ -0,0 +1,11 @@
|
||||
{
|
||||
"workbench.colorCustomizations": {
|
||||
"tab.activeBackground": "#65952acc"
|
||||
},
|
||||
"editor.defaultFormatter": "esbenp.prettier-vscode",
|
||||
"prettier.printWidth": 1024,
|
||||
"prettier.tabWidth": 4,
|
||||
"files.associations": {
|
||||
"*.css": "postcss"
|
||||
}
|
||||
}
|
||||
9
client/demo/.vscode/settings.json
vendored
9
client/demo/.vscode/settings.json
vendored
@ -1,8 +1,11 @@
|
||||
{
|
||||
"files.associations": {
|
||||
"*.css": "postcss"
|
||||
},
|
||||
"workbench.colorCustomizations": {
|
||||
"tab.activeBackground": "#65952acc"
|
||||
},
|
||||
"editor.defaultFormatter": "esbenp.prettier-vscode",
|
||||
"prettier.printWidth": 1024,
|
||||
"prettier.tabWidth": 4,
|
||||
"files.associations": {
|
||||
"*.css": "postcss"
|
||||
}
|
||||
}
|
||||
|
||||
@ -1,4 +1,4 @@
|
||||
import React from "react"
|
||||
import React from "react";
|
||||
import { GuiStateProvider } from "./001_GuiStateProvider";
|
||||
import { Dialogs } from "./900_Dialogs";
|
||||
import { ModelSlotControl } from "./b00_ModelSlotControl";
|
||||
@ -13,5 +13,5 @@ export const Demo = () => {
|
||||
<ModelSlotControl></ModelSlotControl>
|
||||
</div>
|
||||
</GuiStateProvider>
|
||||
)
|
||||
}
|
||||
);
|
||||
};
|
||||
|
||||
15
client/lib/.vscode/settings.json
vendored
15
client/lib/.vscode/settings.json
vendored
@ -1,8 +1,11 @@
|
||||
{
|
||||
"workbench.colorCustomizations": {
|
||||
"tab.activeBackground": "#65952acc"
|
||||
},
|
||||
"editor.defaultFormatter": "esbenp.prettier-vscode",
|
||||
"prettier.printWidth": 1024,
|
||||
"prettier.tabWidth": 4
|
||||
"workbench.colorCustomizations": {
|
||||
"tab.activeBackground": "#65952acc"
|
||||
},
|
||||
"editor.defaultFormatter": "esbenp.prettier-vscode",
|
||||
"prettier.printWidth": 1024,
|
||||
"prettier.tabWidth": 4,
|
||||
"files.associations": {
|
||||
"*.css": "postcss"
|
||||
}
|
||||
}
|
||||
|
||||
@ -4,22 +4,21 @@
|
||||
// 187.5chunk -> 1sec
|
||||
|
||||
export const VoiceChangerType = {
|
||||
MMVCv15: "MMVCv15",
|
||||
MMVCv13: "MMVCv13",
|
||||
"so-vits-svc-40": "so-vits-svc-40",
|
||||
"DDSP-SVC": "DDSP-SVC",
|
||||
RVC: "RVC",
|
||||
"Diffusion-SVC": "Diffusion-SVC",
|
||||
Beatrice: "Beatrice",
|
||||
LLVC: "LLVC",
|
||||
WebModel: "WebModel",
|
||||
EasyVC: "EasyVC",
|
||||
MMVCv15: "MMVCv15",
|
||||
MMVCv13: "MMVCv13",
|
||||
"so-vits-svc-40": "so-vits-svc-40",
|
||||
"DDSP-SVC": "DDSP-SVC",
|
||||
RVC: "RVC",
|
||||
"Diffusion-SVC": "Diffusion-SVC",
|
||||
Beatrice: "Beatrice",
|
||||
LLVC: "LLVC",
|
||||
WebModel: "WebModel",
|
||||
EasyVC: "EasyVC",
|
||||
} as const;
|
||||
export type VoiceChangerType =
|
||||
(typeof VoiceChangerType)[keyof typeof VoiceChangerType];
|
||||
export type VoiceChangerType = (typeof VoiceChangerType)[keyof typeof VoiceChangerType];
|
||||
|
||||
export const StaticModel = {
|
||||
BeatriceJVS: "Beatrice-JVS",
|
||||
BeatriceJVS: "Beatrice-JVS",
|
||||
} as const;
|
||||
export type StaticModel = (typeof StaticModel)[keyof typeof StaticModel];
|
||||
|
||||
@ -27,439 +26,426 @@ export type StaticModel = (typeof StaticModel)[keyof typeof StaticModel];
|
||||
// サーバセッティング
|
||||
///////////////////////
|
||||
export const InputSampleRate = {
|
||||
"48000": 48000,
|
||||
"44100": 44100,
|
||||
"24000": 24000,
|
||||
"48000": 48000,
|
||||
"44100": 44100,
|
||||
"24000": 24000,
|
||||
} as const;
|
||||
export type InputSampleRate =
|
||||
(typeof InputSampleRate)[keyof typeof InputSampleRate];
|
||||
export type InputSampleRate = (typeof InputSampleRate)[keyof typeof InputSampleRate];
|
||||
|
||||
export const ModelSamplingRate = {
|
||||
"48000": 48000,
|
||||
"40000": 40000,
|
||||
"32000": 32000,
|
||||
"48000": 48000,
|
||||
"40000": 40000,
|
||||
"32000": 32000,
|
||||
} as const;
|
||||
export type ModelSamplingRate =
|
||||
(typeof InputSampleRate)[keyof typeof InputSampleRate];
|
||||
export type ModelSamplingRate = (typeof InputSampleRate)[keyof typeof InputSampleRate];
|
||||
|
||||
export const CrossFadeOverlapSize = {
|
||||
"128": 128,
|
||||
"256": 256,
|
||||
"512": 512,
|
||||
"1024": 1024,
|
||||
"2048": 2048,
|
||||
"4096": 4096,
|
||||
"128": 128,
|
||||
"256": 256,
|
||||
"512": 512,
|
||||
"1024": 1024,
|
||||
"2048": 2048,
|
||||
"4096": 4096,
|
||||
} as const;
|
||||
export type CrossFadeOverlapSize =
|
||||
(typeof CrossFadeOverlapSize)[keyof typeof CrossFadeOverlapSize];
|
||||
export type CrossFadeOverlapSize = (typeof CrossFadeOverlapSize)[keyof typeof CrossFadeOverlapSize];
|
||||
|
||||
export const F0Detector = {
|
||||
dio: "dio",
|
||||
harvest: "harvest",
|
||||
crepe: "crepe",
|
||||
crepe_full: "crepe_full",
|
||||
crepe_tiny: "crepe_tiny",
|
||||
rmvpe: "rmvpe",
|
||||
rmvpe_onnx: "rmvpe_onnx",
|
||||
fcpe: "fcpe",
|
||||
dio: "dio",
|
||||
harvest: "harvest",
|
||||
crepe: "crepe",
|
||||
crepe_full: "crepe_full",
|
||||
crepe_tiny: "crepe_tiny",
|
||||
rmvpe: "rmvpe",
|
||||
rmvpe_onnx: "rmvpe_onnx",
|
||||
fcpe: "fcpe",
|
||||
} as const;
|
||||
export type F0Detector = (typeof F0Detector)[keyof typeof F0Detector];
|
||||
|
||||
export const DiffMethod = {
|
||||
pndm: "pndm",
|
||||
"dpm-solver": "dpm-solver",
|
||||
pndm: "pndm",
|
||||
"dpm-solver": "dpm-solver",
|
||||
} as const;
|
||||
export type DiffMethod = (typeof DiffMethod)[keyof typeof DiffMethod];
|
||||
|
||||
export const RVCModelType = {
|
||||
pyTorchRVC: "pyTorchRVC",
|
||||
pyTorchRVCNono: "pyTorchRVCNono",
|
||||
pyTorchRVCv2: "pyTorchRVCv2",
|
||||
pyTorchRVCv2Nono: "pyTorchRVCv2Nono",
|
||||
pyTorchWebUI: "pyTorchWebUI",
|
||||
pyTorchWebUINono: "pyTorchWebUINono",
|
||||
onnxRVC: "onnxRVC",
|
||||
onnxRVCNono: "onnxRVCNono",
|
||||
pyTorchRVC: "pyTorchRVC",
|
||||
pyTorchRVCNono: "pyTorchRVCNono",
|
||||
pyTorchRVCv2: "pyTorchRVCv2",
|
||||
pyTorchRVCv2Nono: "pyTorchRVCv2Nono",
|
||||
pyTorchWebUI: "pyTorchWebUI",
|
||||
pyTorchWebUINono: "pyTorchWebUINono",
|
||||
onnxRVC: "onnxRVC",
|
||||
onnxRVCNono: "onnxRVCNono",
|
||||
} as const;
|
||||
export type RVCModelType = (typeof RVCModelType)[keyof typeof RVCModelType];
|
||||
|
||||
export const ServerSettingKey = {
|
||||
passThrough: "passThrough",
|
||||
srcId: "srcId",
|
||||
dstId: "dstId",
|
||||
gpu: "gpu",
|
||||
passThrough: "passThrough",
|
||||
srcId: "srcId",
|
||||
dstId: "dstId",
|
||||
gpu: "gpu",
|
||||
|
||||
crossFadeOffsetRate: "crossFadeOffsetRate",
|
||||
crossFadeEndRate: "crossFadeEndRate",
|
||||
crossFadeOverlapSize: "crossFadeOverlapSize",
|
||||
crossFadeOffsetRate: "crossFadeOffsetRate",
|
||||
crossFadeEndRate: "crossFadeEndRate",
|
||||
crossFadeOverlapSize: "crossFadeOverlapSize",
|
||||
|
||||
framework: "framework",
|
||||
onnxExecutionProvider: "onnxExecutionProvider",
|
||||
framework: "framework",
|
||||
onnxExecutionProvider: "onnxExecutionProvider",
|
||||
|
||||
f0Factor: "f0Factor",
|
||||
f0Detector: "f0Detector",
|
||||
recordIO: "recordIO",
|
||||
f0Factor: "f0Factor",
|
||||
f0Detector: "f0Detector",
|
||||
recordIO: "recordIO",
|
||||
|
||||
enableServerAudio: "enableServerAudio",
|
||||
serverAudioStated: "serverAudioStated",
|
||||
serverAudioSampleRate: "serverAudioSampleRate",
|
||||
serverInputAudioSampleRate: "serverInputAudioSampleRate",
|
||||
serverOutputAudioSampleRate: "serverOutputAudioSampleRate",
|
||||
serverMonitorAudioSampleRate: "serverMonitorAudioSampleRate",
|
||||
serverInputAudioBufferSize: "serverInputAudioBufferSize",
|
||||
serverOutputAudioBufferSize: "serverOutputAudioBufferSize",
|
||||
serverInputDeviceId: "serverInputDeviceId",
|
||||
serverOutputDeviceId: "serverOutputDeviceId",
|
||||
serverMonitorDeviceId: "serverMonitorDeviceId",
|
||||
serverReadChunkSize: "serverReadChunkSize",
|
||||
serverInputAudioGain: "serverInputAudioGain",
|
||||
serverOutputAudioGain: "serverOutputAudioGain",
|
||||
serverMonitorAudioGain: "serverMonitorAudioGain",
|
||||
enableServerAudio: "enableServerAudio",
|
||||
serverAudioStated: "serverAudioStated",
|
||||
serverAudioSampleRate: "serverAudioSampleRate",
|
||||
serverInputAudioSampleRate: "serverInputAudioSampleRate",
|
||||
serverOutputAudioSampleRate: "serverOutputAudioSampleRate",
|
||||
serverMonitorAudioSampleRate: "serverMonitorAudioSampleRate",
|
||||
serverInputAudioBufferSize: "serverInputAudioBufferSize",
|
||||
serverOutputAudioBufferSize: "serverOutputAudioBufferSize",
|
||||
serverInputDeviceId: "serverInputDeviceId",
|
||||
serverOutputDeviceId: "serverOutputDeviceId",
|
||||
serverMonitorDeviceId: "serverMonitorDeviceId",
|
||||
serverReadChunkSize: "serverReadChunkSize",
|
||||
serverInputAudioGain: "serverInputAudioGain",
|
||||
serverOutputAudioGain: "serverOutputAudioGain",
|
||||
serverMonitorAudioGain: "serverMonitorAudioGain",
|
||||
|
||||
tran: "tran",
|
||||
noiseScale: "noiseScale",
|
||||
predictF0: "predictF0",
|
||||
silentThreshold: "silentThreshold",
|
||||
extraConvertSize: "extraConvertSize",
|
||||
clusterInferRatio: "clusterInferRatio",
|
||||
tran: "tran",
|
||||
noiseScale: "noiseScale",
|
||||
predictF0: "predictF0",
|
||||
silentThreshold: "silentThreshold",
|
||||
extraConvertSize: "extraConvertSize",
|
||||
clusterInferRatio: "clusterInferRatio",
|
||||
|
||||
indexRatio: "indexRatio",
|
||||
protect: "protect",
|
||||
rvcQuality: "rvcQuality",
|
||||
modelSamplingRate: "modelSamplingRate",
|
||||
silenceFront: "silenceFront",
|
||||
modelSlotIndex: "modelSlotIndex",
|
||||
indexRatio: "indexRatio",
|
||||
protect: "protect",
|
||||
rvcQuality: "rvcQuality",
|
||||
modelSamplingRate: "modelSamplingRate",
|
||||
silenceFront: "silenceFront",
|
||||
modelSlotIndex: "modelSlotIndex",
|
||||
|
||||
useEnhancer: "useEnhancer",
|
||||
useDiff: "useDiff",
|
||||
// "useDiffDpm": "useDiffDpm",
|
||||
diffMethod: "diffMethod",
|
||||
useDiffSilence: "useDiffSilence",
|
||||
diffAcc: "diffAcc",
|
||||
diffSpkId: "diffSpkId",
|
||||
kStep: "kStep",
|
||||
threshold: "threshold",
|
||||
useEnhancer: "useEnhancer",
|
||||
useDiff: "useDiff",
|
||||
// "useDiffDpm": "useDiffDpm",
|
||||
diffMethod: "diffMethod",
|
||||
useDiffSilence: "useDiffSilence",
|
||||
diffAcc: "diffAcc",
|
||||
diffSpkId: "diffSpkId",
|
||||
kStep: "kStep",
|
||||
threshold: "threshold",
|
||||
|
||||
speedUp: "speedUp",
|
||||
skipDiffusion: "skipDiffusion",
|
||||
speedUp: "speedUp",
|
||||
skipDiffusion: "skipDiffusion",
|
||||
|
||||
inputSampleRate: "inputSampleRate",
|
||||
enableDirectML: "enableDirectML",
|
||||
inputSampleRate: "inputSampleRate",
|
||||
enableDirectML: "enableDirectML",
|
||||
} as const;
|
||||
export type ServerSettingKey =
|
||||
(typeof ServerSettingKey)[keyof typeof ServerSettingKey];
|
||||
export type ServerSettingKey = (typeof ServerSettingKey)[keyof typeof ServerSettingKey];
|
||||
|
||||
export type VoiceChangerServerSetting = {
|
||||
passThrough: boolean;
|
||||
srcId: number;
|
||||
dstId: number;
|
||||
gpu: number;
|
||||
passThrough: boolean;
|
||||
srcId: number;
|
||||
dstId: number;
|
||||
gpu: number;
|
||||
|
||||
crossFadeOffsetRate: number;
|
||||
crossFadeEndRate: number;
|
||||
crossFadeOverlapSize: CrossFadeOverlapSize;
|
||||
crossFadeOffsetRate: number;
|
||||
crossFadeEndRate: number;
|
||||
crossFadeOverlapSize: CrossFadeOverlapSize;
|
||||
|
||||
f0Factor: number;
|
||||
f0Detector: F0Detector; // dio or harvest
|
||||
recordIO: number; // 0:off, 1:on
|
||||
f0Factor: number;
|
||||
f0Detector: F0Detector; // dio or harvest
|
||||
recordIO: number; // 0:off, 1:on
|
||||
|
||||
enableServerAudio: number; // 0:off, 1:on
|
||||
serverAudioStated: number; // 0:off, 1:on
|
||||
serverAudioSampleRate: number;
|
||||
serverInputAudioSampleRate: number;
|
||||
serverOutputAudioSampleRate: number;
|
||||
serverMonitorAudioSampleRate: number;
|
||||
serverInputAudioBufferSize: number;
|
||||
serverOutputAudioBufferSize: number;
|
||||
serverInputDeviceId: number;
|
||||
serverOutputDeviceId: number;
|
||||
serverMonitorDeviceId: number;
|
||||
serverReadChunkSize: number;
|
||||
serverInputAudioGain: number;
|
||||
serverOutputAudioGain: number;
|
||||
serverMonitorAudioGain: number;
|
||||
enableServerAudio: number; // 0:off, 1:on
|
||||
serverAudioStated: number; // 0:off, 1:on
|
||||
serverAudioSampleRate: number;
|
||||
serverInputAudioSampleRate: number;
|
||||
serverOutputAudioSampleRate: number;
|
||||
serverMonitorAudioSampleRate: number;
|
||||
serverInputAudioBufferSize: number;
|
||||
serverOutputAudioBufferSize: number;
|
||||
serverInputDeviceId: number;
|
||||
serverOutputDeviceId: number;
|
||||
serverMonitorDeviceId: number;
|
||||
serverReadChunkSize: number;
|
||||
serverInputAudioGain: number;
|
||||
serverOutputAudioGain: number;
|
||||
serverMonitorAudioGain: number;
|
||||
|
||||
tran: number; // so-vits-svc
|
||||
noiseScale: number; // so-vits-svc
|
||||
predictF0: number; // so-vits-svc
|
||||
silentThreshold: number; // so-vits-svc
|
||||
extraConvertSize: number; // so-vits-svc
|
||||
clusterInferRatio: number; // so-vits-svc
|
||||
tran: number; // so-vits-svc
|
||||
noiseScale: number; // so-vits-svc
|
||||
predictF0: number; // so-vits-svc
|
||||
silentThreshold: number; // so-vits-svc
|
||||
extraConvertSize: number; // so-vits-svc
|
||||
clusterInferRatio: number; // so-vits-svc
|
||||
|
||||
indexRatio: number; // RVC
|
||||
protect: number; // RVC
|
||||
rvcQuality: number; // 0:low, 1:high
|
||||
silenceFront: number; // 0:off, 1:on
|
||||
modelSamplingRate: ModelSamplingRate; // 32000,40000,48000
|
||||
modelSlotIndex: number | StaticModel;
|
||||
indexRatio: number; // RVC
|
||||
protect: number; // RVC
|
||||
rvcQuality: number; // 0:low, 1:high
|
||||
silenceFront: number; // 0:off, 1:on
|
||||
modelSamplingRate: ModelSamplingRate; // 32000,40000,48000
|
||||
modelSlotIndex: number | StaticModel;
|
||||
|
||||
useEnhancer: number; // DDSP-SVC
|
||||
useDiff: number; // DDSP-SVC
|
||||
// useDiffDpm: number// DDSP-SVC
|
||||
diffMethod: DiffMethod; // DDSP-SVC
|
||||
useDiffSilence: number; // DDSP-SVC
|
||||
diffAcc: number; // DDSP-SVC
|
||||
diffSpkId: number; // DDSP-SVC
|
||||
kStep: number; // DDSP-SVC
|
||||
threshold: number; // DDSP-SVC
|
||||
useEnhancer: number; // DDSP-SVC
|
||||
useDiff: number; // DDSP-SVC
|
||||
// useDiffDpm: number// DDSP-SVC
|
||||
diffMethod: DiffMethod; // DDSP-SVC
|
||||
useDiffSilence: number; // DDSP-SVC
|
||||
diffAcc: number; // DDSP-SVC
|
||||
diffSpkId: number; // DDSP-SVC
|
||||
kStep: number; // DDSP-SVC
|
||||
threshold: number; // DDSP-SVC
|
||||
|
||||
speedUp: number; // Diffusion-SVC
|
||||
skipDiffusion: number; // Diffusion-SVC 0:off, 1:on
|
||||
speedUp: number; // Diffusion-SVC
|
||||
skipDiffusion: number; // Diffusion-SVC 0:off, 1:on
|
||||
|
||||
inputSampleRate: InputSampleRate;
|
||||
enableDirectML: number;
|
||||
inputSampleRate: InputSampleRate;
|
||||
enableDirectML: number;
|
||||
};
|
||||
|
||||
type ModelSlot = {
|
||||
slotIndex: number | StaticModel;
|
||||
voiceChangerType: VoiceChangerType;
|
||||
name: string;
|
||||
description: string;
|
||||
credit: string;
|
||||
termsOfUseUrl: string;
|
||||
iconFile: string;
|
||||
speakers: { [key: number]: string };
|
||||
slotIndex: number | StaticModel;
|
||||
voiceChangerType: VoiceChangerType;
|
||||
name: string;
|
||||
description: string;
|
||||
credit: string;
|
||||
termsOfUseUrl: string;
|
||||
iconFile: string;
|
||||
speakers: { [key: number]: string };
|
||||
};
|
||||
|
||||
export type RVCModelSlot = ModelSlot & {
|
||||
modelFile: string;
|
||||
indexFile: string;
|
||||
defaultIndexRatio: number;
|
||||
defaultProtect: number;
|
||||
defaultTune: number;
|
||||
modelType: RVCModelType;
|
||||
modelFile: string;
|
||||
indexFile: string;
|
||||
defaultIndexRatio: number;
|
||||
defaultProtect: number;
|
||||
defaultTune: number;
|
||||
modelType: RVCModelType;
|
||||
|
||||
embChannels: number;
|
||||
f0: boolean;
|
||||
samplingRate: number;
|
||||
deprecated: boolean;
|
||||
embChannels: number;
|
||||
f0: boolean;
|
||||
samplingRate: number;
|
||||
deprecated: boolean;
|
||||
};
|
||||
|
||||
export type MMVCv13ModelSlot = ModelSlot & {
|
||||
modelFile: string;
|
||||
configFile: string;
|
||||
srcId: number;
|
||||
dstId: number;
|
||||
modelFile: string;
|
||||
configFile: string;
|
||||
srcId: number;
|
||||
dstId: number;
|
||||
|
||||
samplingRate: number;
|
||||
speakers: { [key: number]: string };
|
||||
samplingRate: number;
|
||||
speakers: { [key: number]: string };
|
||||
};
|
||||
|
||||
export type MMVCv15ModelSlot = ModelSlot & {
|
||||
modelFile: string;
|
||||
configFile: string;
|
||||
srcId: number;
|
||||
dstId: number;
|
||||
f0Factor: number;
|
||||
samplingRate: number;
|
||||
f0: { [key: number]: number };
|
||||
modelFile: string;
|
||||
configFile: string;
|
||||
srcId: number;
|
||||
dstId: number;
|
||||
f0Factor: number;
|
||||
samplingRate: number;
|
||||
f0: { [key: number]: number };
|
||||
};
|
||||
|
||||
export type SoVitsSvc40ModelSlot = ModelSlot & {
|
||||
modelFile: string;
|
||||
configFile: string;
|
||||
clusterFile: string;
|
||||
dstId: number;
|
||||
modelFile: string;
|
||||
configFile: string;
|
||||
clusterFile: string;
|
||||
dstId: number;
|
||||
|
||||
samplingRate: number;
|
||||
samplingRate: number;
|
||||
|
||||
defaultTune: number;
|
||||
defaultClusterInferRatio: number;
|
||||
noiseScale: number;
|
||||
speakers: { [key: number]: string };
|
||||
defaultTune: number;
|
||||
defaultClusterInferRatio: number;
|
||||
noiseScale: number;
|
||||
speakers: { [key: number]: string };
|
||||
};
|
||||
|
||||
export type DDSPSVCModelSlot = ModelSlot & {
|
||||
modelFile: string;
|
||||
configFile: string;
|
||||
diffModelFile: string;
|
||||
diffConfigFile: string;
|
||||
dstId: number;
|
||||
modelFile: string;
|
||||
configFile: string;
|
||||
diffModelFile: string;
|
||||
diffConfigFile: string;
|
||||
dstId: number;
|
||||
|
||||
samplingRate: number;
|
||||
samplingRate: number;
|
||||
|
||||
defaultTune: number;
|
||||
enhancer: boolean;
|
||||
diffusion: boolean;
|
||||
acc: number;
|
||||
kstep: number;
|
||||
speakers: { [key: number]: string };
|
||||
defaultTune: number;
|
||||
enhancer: boolean;
|
||||
diffusion: boolean;
|
||||
acc: number;
|
||||
kstep: number;
|
||||
speakers: { [key: number]: string };
|
||||
};
|
||||
|
||||
export type DiffusionSVCModelSlot = ModelSlot & {
|
||||
modelFile: string;
|
||||
dstId: number;
|
||||
modelFile: string;
|
||||
dstId: number;
|
||||
|
||||
samplingRate: number;
|
||||
samplingRate: number;
|
||||
|
||||
defaultTune: number;
|
||||
defaultKstep: number;
|
||||
defaultSpeedup: number;
|
||||
kStepMax: number;
|
||||
nLayers: number;
|
||||
nnLayers: number;
|
||||
speakers: { [key: number]: string };
|
||||
defaultTune: number;
|
||||
defaultKstep: number;
|
||||
defaultSpeedup: number;
|
||||
kStepMax: number;
|
||||
nLayers: number;
|
||||
nnLayers: number;
|
||||
speakers: { [key: number]: string };
|
||||
};
|
||||
|
||||
export type BeatriceModelSlot = ModelSlot & {
|
||||
modelFile: string;
|
||||
dstId: number;
|
||||
modelFile: string;
|
||||
dstId: number;
|
||||
|
||||
speakers: { [key: number]: string };
|
||||
speakers: { [key: number]: string };
|
||||
};
|
||||
|
||||
export type LLVCModelSlot = ModelSlot & {
|
||||
modelFile: string;
|
||||
configFile: string;
|
||||
modelFile: string;
|
||||
configFile: string;
|
||||
|
||||
speakers: { [key: number]: string };
|
||||
speakers: { [key: number]: string };
|
||||
};
|
||||
|
||||
export type WebModelSlot = ModelSlot & {
|
||||
modelFile: string;
|
||||
defaultTune: number;
|
||||
modelType: RVCModelType;
|
||||
f0: boolean;
|
||||
samplingRate: number;
|
||||
modelFile: string;
|
||||
defaultTune: number;
|
||||
modelType: RVCModelType;
|
||||
f0: boolean;
|
||||
samplingRate: number;
|
||||
};
|
||||
|
||||
export type ModelSlotUnion =
|
||||
| RVCModelSlot
|
||||
| MMVCv13ModelSlot
|
||||
| MMVCv15ModelSlot
|
||||
| SoVitsSvc40ModelSlot
|
||||
| DDSPSVCModelSlot
|
||||
| DiffusionSVCModelSlot
|
||||
| BeatriceModelSlot
|
||||
| LLVCModelSlot
|
||||
| WebModelSlot;
|
||||
export type ModelSlotUnion = RVCModelSlot | MMVCv13ModelSlot | MMVCv15ModelSlot | SoVitsSvc40ModelSlot | DDSPSVCModelSlot | DiffusionSVCModelSlot | BeatriceModelSlot | LLVCModelSlot | WebModelSlot;
|
||||
|
||||
type ServerAudioDevice = {
|
||||
kind: "audioinput" | "audiooutput";
|
||||
index: number;
|
||||
name: string;
|
||||
hostAPI: string;
|
||||
kind: "audioinput" | "audiooutput";
|
||||
index: number;
|
||||
name: string;
|
||||
hostAPI: string;
|
||||
};
|
||||
|
||||
export type ServerInfo = VoiceChangerServerSetting & {
|
||||
// コンフィグ対象外 (getInfoで取得のみ可能な情報)
|
||||
status: string;
|
||||
modelSlots: ModelSlotUnion[];
|
||||
serverAudioInputDevices: ServerAudioDevice[];
|
||||
serverAudioOutputDevices: ServerAudioDevice[];
|
||||
sampleModels: (RVCSampleModel | DiffusionSVCSampleModel)[];
|
||||
gpus: {
|
||||
id: number;
|
||||
name: string;
|
||||
memory: number;
|
||||
}[];
|
||||
maxInputLength: number; // MMVCv15
|
||||
voiceChangerParams: {
|
||||
model_dir: string;
|
||||
};
|
||||
// コンフィグ対象外 (getInfoで取得のみ可能な情報)
|
||||
status: string;
|
||||
modelSlots: ModelSlotUnion[];
|
||||
serverAudioInputDevices: ServerAudioDevice[];
|
||||
serverAudioOutputDevices: ServerAudioDevice[];
|
||||
sampleModels: (RVCSampleModel | DiffusionSVCSampleModel)[];
|
||||
gpus: {
|
||||
id: number;
|
||||
name: string;
|
||||
memory: number;
|
||||
}[];
|
||||
maxInputLength: number; // MMVCv15
|
||||
voiceChangerParams: {
|
||||
model_dir: string;
|
||||
};
|
||||
};
|
||||
|
||||
export type SampleModel = {
|
||||
id: string;
|
||||
voiceChangerType: VoiceChangerType;
|
||||
lang: string;
|
||||
tag: string[];
|
||||
name: string;
|
||||
modelUrl: string;
|
||||
termsOfUseUrl: string;
|
||||
icon: string;
|
||||
credit: string;
|
||||
description: string;
|
||||
sampleRate: number;
|
||||
modelType: string;
|
||||
f0: boolean;
|
||||
id: string;
|
||||
voiceChangerType: VoiceChangerType;
|
||||
lang: string;
|
||||
tag: string[];
|
||||
name: string;
|
||||
modelUrl: string;
|
||||
termsOfUseUrl: string;
|
||||
icon: string;
|
||||
credit: string;
|
||||
description: string;
|
||||
sampleRate: number;
|
||||
modelType: string;
|
||||
f0: boolean;
|
||||
};
|
||||
|
||||
export type RVCSampleModel = SampleModel & {
|
||||
indexUrl: string;
|
||||
featureUrl: string;
|
||||
indexUrl: string;
|
||||
featureUrl: string;
|
||||
};
|
||||
|
||||
export type DiffusionSVCSampleModel = SampleModel & {
|
||||
numOfDiffLayers: number;
|
||||
numOfNativeLayers: number;
|
||||
maxKStep: number;
|
||||
numOfDiffLayers: number;
|
||||
numOfNativeLayers: number;
|
||||
maxKStep: number;
|
||||
};
|
||||
|
||||
export const DefaultServerSetting: ServerInfo = {
|
||||
// VC Common
|
||||
passThrough: false,
|
||||
inputSampleRate: 48000,
|
||||
// VC Common
|
||||
passThrough: false,
|
||||
inputSampleRate: 48000,
|
||||
|
||||
crossFadeOffsetRate: 0.0,
|
||||
crossFadeEndRate: 1.0,
|
||||
crossFadeOverlapSize: CrossFadeOverlapSize[1024],
|
||||
crossFadeOffsetRate: 0.0,
|
||||
crossFadeEndRate: 1.0,
|
||||
crossFadeOverlapSize: CrossFadeOverlapSize[1024],
|
||||
|
||||
recordIO: 0,
|
||||
recordIO: 0,
|
||||
|
||||
enableServerAudio: 0,
|
||||
serverAudioStated: 0,
|
||||
serverAudioSampleRate: 48000,
|
||||
serverInputAudioSampleRate: 48000,
|
||||
serverOutputAudioSampleRate: 48000,
|
||||
serverMonitorAudioSampleRate: 48000,
|
||||
serverInputAudioBufferSize: 1024 * 24,
|
||||
serverOutputAudioBufferSize: 1024 * 24,
|
||||
serverInputDeviceId: -1,
|
||||
serverOutputDeviceId: -1,
|
||||
serverMonitorDeviceId: -1,
|
||||
serverReadChunkSize: 256,
|
||||
serverInputAudioGain: 1.0,
|
||||
serverOutputAudioGain: 1.0,
|
||||
serverMonitorAudioGain: 1.0,
|
||||
enableServerAudio: 0,
|
||||
serverAudioStated: 0,
|
||||
serverAudioSampleRate: 48000,
|
||||
serverInputAudioSampleRate: 48000,
|
||||
serverOutputAudioSampleRate: 48000,
|
||||
serverMonitorAudioSampleRate: 48000,
|
||||
serverInputAudioBufferSize: 1024 * 24,
|
||||
serverOutputAudioBufferSize: 1024 * 24,
|
||||
serverInputDeviceId: -1,
|
||||
serverOutputDeviceId: -1,
|
||||
serverMonitorDeviceId: -1,
|
||||
serverReadChunkSize: 256,
|
||||
serverInputAudioGain: 1.0,
|
||||
serverOutputAudioGain: 1.0,
|
||||
serverMonitorAudioGain: 1.0,
|
||||
|
||||
// VC Specific
|
||||
srcId: 0,
|
||||
dstId: 1,
|
||||
gpu: 0,
|
||||
// VC Specific
|
||||
srcId: 0,
|
||||
dstId: 1,
|
||||
gpu: 0,
|
||||
|
||||
f0Factor: 1.0,
|
||||
f0Detector: F0Detector.rmvpe_onnx,
|
||||
f0Factor: 1.0,
|
||||
f0Detector: F0Detector.rmvpe_onnx,
|
||||
|
||||
tran: 0,
|
||||
noiseScale: 0,
|
||||
predictF0: 0,
|
||||
silentThreshold: 0,
|
||||
extraConvertSize: 0,
|
||||
clusterInferRatio: 0,
|
||||
tran: 0,
|
||||
noiseScale: 0,
|
||||
predictF0: 0,
|
||||
silentThreshold: 0,
|
||||
extraConvertSize: 0,
|
||||
clusterInferRatio: 0,
|
||||
|
||||
indexRatio: 0,
|
||||
protect: 0.5,
|
||||
rvcQuality: 0,
|
||||
modelSamplingRate: 48000,
|
||||
silenceFront: 1,
|
||||
modelSlotIndex: 0,
|
||||
sampleModels: [],
|
||||
gpus: [],
|
||||
indexRatio: 0,
|
||||
protect: 0.5,
|
||||
rvcQuality: 0,
|
||||
modelSamplingRate: 48000,
|
||||
silenceFront: 1,
|
||||
modelSlotIndex: 0,
|
||||
sampleModels: [],
|
||||
gpus: [],
|
||||
|
||||
useEnhancer: 0,
|
||||
useDiff: 1,
|
||||
diffMethod: "dpm-solver",
|
||||
useDiffSilence: 0,
|
||||
diffAcc: 20,
|
||||
diffSpkId: 1,
|
||||
kStep: 120,
|
||||
threshold: -45,
|
||||
useEnhancer: 0,
|
||||
useDiff: 1,
|
||||
diffMethod: "dpm-solver",
|
||||
useDiffSilence: 0,
|
||||
diffAcc: 20,
|
||||
diffSpkId: 1,
|
||||
kStep: 120,
|
||||
threshold: -45,
|
||||
|
||||
speedUp: 10,
|
||||
skipDiffusion: 1,
|
||||
speedUp: 10,
|
||||
skipDiffusion: 1,
|
||||
|
||||
enableDirectML: 0,
|
||||
//
|
||||
status: "ok",
|
||||
modelSlots: [],
|
||||
serverAudioInputDevices: [],
|
||||
serverAudioOutputDevices: [],
|
||||
enableDirectML: 0,
|
||||
//
|
||||
status: "ok",
|
||||
modelSlots: [],
|
||||
serverAudioInputDevices: [],
|
||||
serverAudioOutputDevices: [],
|
||||
|
||||
maxInputLength: 128 * 2048,
|
||||
voiceChangerParams: {
|
||||
model_dir: "",
|
||||
},
|
||||
maxInputLength: 128 * 2048,
|
||||
voiceChangerParams: {
|
||||
model_dir: "",
|
||||
},
|
||||
};
|
||||
|
||||
///////////////////////
|
||||
@ -467,113 +453,109 @@ export const DefaultServerSetting: ServerInfo = {
|
||||
///////////////////////
|
||||
|
||||
export type WorkletSetting = {
|
||||
numTrancateTreshold: number;
|
||||
volTrancateThreshold: number;
|
||||
volTrancateLength: number;
|
||||
numTrancateTreshold: number;
|
||||
volTrancateThreshold: number;
|
||||
volTrancateLength: number;
|
||||
};
|
||||
///////////////////////
|
||||
// Worklet Nodeセッティング
|
||||
///////////////////////
|
||||
export const Protocol = {
|
||||
sio: "sio",
|
||||
rest: "rest",
|
||||
internal: "internal",
|
||||
sio: "sio",
|
||||
rest: "rest",
|
||||
internal: "internal",
|
||||
} as const;
|
||||
export type Protocol = (typeof Protocol)[keyof typeof Protocol];
|
||||
|
||||
export const SendingSampleRate = {
|
||||
"48000": 48000,
|
||||
"44100": 44100,
|
||||
"24000": 24000,
|
||||
"48000": 48000,
|
||||
"44100": 44100,
|
||||
"24000": 24000,
|
||||
} as const;
|
||||
export type SendingSampleRate =
|
||||
(typeof SendingSampleRate)[keyof typeof SendingSampleRate];
|
||||
export type SendingSampleRate = (typeof SendingSampleRate)[keyof typeof SendingSampleRate];
|
||||
|
||||
export const DownSamplingMode = {
|
||||
decimate: "decimate",
|
||||
average: "average",
|
||||
decimate: "decimate",
|
||||
average: "average",
|
||||
} as const;
|
||||
export type DownSamplingMode =
|
||||
(typeof DownSamplingMode)[keyof typeof DownSamplingMode];
|
||||
export type DownSamplingMode = (typeof DownSamplingMode)[keyof typeof DownSamplingMode];
|
||||
|
||||
export type WorkletNodeSetting = {
|
||||
serverUrl: string;
|
||||
protocol: Protocol;
|
||||
sendingSampleRate: SendingSampleRate;
|
||||
inputChunkNum: number;
|
||||
downSamplingMode: DownSamplingMode;
|
||||
serverUrl: string;
|
||||
protocol: Protocol;
|
||||
sendingSampleRate: SendingSampleRate;
|
||||
inputChunkNum: number;
|
||||
downSamplingMode: DownSamplingMode;
|
||||
};
|
||||
|
||||
///////////////////////
|
||||
// クライアントセッティング
|
||||
///////////////////////
|
||||
export const SampleRate = {
|
||||
"48000": 48000,
|
||||
"48000": 48000,
|
||||
} as const;
|
||||
export type SampleRate = (typeof SampleRate)[keyof typeof SampleRate];
|
||||
|
||||
export type VoiceChangerClientSetting = {
|
||||
audioInput: string | MediaStream | null;
|
||||
sampleRate: SampleRate; // 48000Hz
|
||||
echoCancel: boolean;
|
||||
noiseSuppression: boolean;
|
||||
noiseSuppression2: boolean;
|
||||
audioInput: string | MediaStream | null;
|
||||
sampleRate: SampleRate; // 48000Hz
|
||||
echoCancel: boolean;
|
||||
noiseSuppression: boolean;
|
||||
noiseSuppression2: boolean;
|
||||
|
||||
inputGain: number;
|
||||
outputGain: number;
|
||||
monitorGain: number;
|
||||
inputGain: number;
|
||||
outputGain: number;
|
||||
monitorGain: number;
|
||||
|
||||
passThroughConfirmationSkip: boolean;
|
||||
passThroughConfirmationSkip: boolean;
|
||||
};
|
||||
|
||||
///////////////////////
|
||||
// Client セッティング
|
||||
///////////////////////
|
||||
export type ClientSetting = {
|
||||
workletSetting: WorkletSetting;
|
||||
workletNodeSetting: WorkletNodeSetting;
|
||||
voiceChangerClientSetting: VoiceChangerClientSetting;
|
||||
workletSetting: WorkletSetting;
|
||||
workletNodeSetting: WorkletNodeSetting;
|
||||
voiceChangerClientSetting: VoiceChangerClientSetting;
|
||||
};
|
||||
export const DefaultClientSettng: ClientSetting = {
|
||||
workletSetting: {
|
||||
// numTrancateTreshold: 512 * 2,
|
||||
numTrancateTreshold: 100,
|
||||
volTrancateThreshold: 0.0005,
|
||||
volTrancateLength: 32,
|
||||
},
|
||||
workletNodeSetting: {
|
||||
serverUrl: "",
|
||||
protocol: "sio",
|
||||
sendingSampleRate: 48000,
|
||||
inputChunkNum: 192,
|
||||
downSamplingMode: "average",
|
||||
},
|
||||
voiceChangerClientSetting: {
|
||||
audioInput: null,
|
||||
sampleRate: 48000,
|
||||
echoCancel: false,
|
||||
noiseSuppression: false,
|
||||
noiseSuppression2: false,
|
||||
inputGain: 1.0,
|
||||
outputGain: 1.0,
|
||||
monitorGain: 1.0,
|
||||
passThroughConfirmationSkip: false,
|
||||
},
|
||||
workletSetting: {
|
||||
// numTrancateTreshold: 512 * 2,
|
||||
numTrancateTreshold: 100,
|
||||
volTrancateThreshold: 0.0005,
|
||||
volTrancateLength: 32,
|
||||
},
|
||||
workletNodeSetting: {
|
||||
serverUrl: "",
|
||||
protocol: "sio",
|
||||
sendingSampleRate: 48000,
|
||||
inputChunkNum: 192,
|
||||
downSamplingMode: "average",
|
||||
},
|
||||
voiceChangerClientSetting: {
|
||||
audioInput: null,
|
||||
sampleRate: 48000,
|
||||
echoCancel: false,
|
||||
noiseSuppression: false,
|
||||
noiseSuppression2: false,
|
||||
inputGain: 1.0,
|
||||
outputGain: 1.0,
|
||||
monitorGain: 1.0,
|
||||
passThroughConfirmationSkip: false,
|
||||
},
|
||||
};
|
||||
|
||||
////////////////////////////////////
|
||||
// Exceptions
|
||||
////////////////////////////////////
|
||||
export const VOICE_CHANGER_CLIENT_EXCEPTION = {
|
||||
ERR_SIO_CONNECT_FAILED: "ERR_SIO_CONNECT_FAILED",
|
||||
ERR_SIO_INVALID_RESPONSE: "ERR_SIO_INVALID_RESPONSE",
|
||||
ERR_REST_INVALID_RESPONSE: "ERR_REST_INVALID_RESPONSE",
|
||||
ERR_MIC_STREAM_NOT_INITIALIZED: "ERR_MIC_STREAM_NOT_INITIALIZED",
|
||||
ERR_INTERNAL_AUDIO_PROCESS_CALLBACK_IS_NOT_INITIALIZED:
|
||||
"ERR_INTERNAL_AUDIO_PROCESS_CALLBACK_IS_NOT_INITIALIZED",
|
||||
ERR_SIO_CONNECT_FAILED: "ERR_SIO_CONNECT_FAILED",
|
||||
ERR_SIO_INVALID_RESPONSE: "ERR_SIO_INVALID_RESPONSE",
|
||||
ERR_REST_INVALID_RESPONSE: "ERR_REST_INVALID_RESPONSE",
|
||||
ERR_MIC_STREAM_NOT_INITIALIZED: "ERR_MIC_STREAM_NOT_INITIALIZED",
|
||||
ERR_INTERNAL_AUDIO_PROCESS_CALLBACK_IS_NOT_INITIALIZED: "ERR_INTERNAL_AUDIO_PROCESS_CALLBACK_IS_NOT_INITIALIZED",
|
||||
} as const;
|
||||
export type VOICE_CHANGER_CLIENT_EXCEPTION =
|
||||
(typeof VOICE_CHANGER_CLIENT_EXCEPTION)[keyof typeof VOICE_CHANGER_CLIENT_EXCEPTION];
|
||||
export type VOICE_CHANGER_CLIENT_EXCEPTION = (typeof VOICE_CHANGER_CLIENT_EXCEPTION)[keyof typeof VOICE_CHANGER_CLIENT_EXCEPTION];
|
||||
|
||||
////////////////////////////////////
|
||||
// indexedDB
|
||||
@ -582,23 +564,22 @@ export const INDEXEDDB_DB_APP_NAME = "INDEXEDDB_KEY_VOICE_CHANGER";
|
||||
export const INDEXEDDB_DB_NAME = "INDEXEDDB_KEY_VOICE_CHANGER_DB";
|
||||
export const INDEXEDDB_KEY_CLIENT = "INDEXEDDB_KEY_VOICE_CHANGER_LIB_CLIENT";
|
||||
export const INDEXEDDB_KEY_SERVER = "INDEXEDDB_KEY_VOICE_CHANGER_LIB_SERVER";
|
||||
export const INDEXEDDB_KEY_MODEL_DATA =
|
||||
"INDEXEDDB_KEY_VOICE_CHANGER_LIB_MODEL_DATA";
|
||||
export const INDEXEDDB_KEY_MODEL_DATA = "INDEXEDDB_KEY_VOICE_CHANGER_LIB_MODEL_DATA";
|
||||
|
||||
// ONNX
|
||||
export type OnnxExporterInfo = {
|
||||
status: string;
|
||||
path: string;
|
||||
filename: string;
|
||||
status: string;
|
||||
path: string;
|
||||
filename: string;
|
||||
};
|
||||
|
||||
// Merge
|
||||
export type MergeElement = {
|
||||
slotIndex: number;
|
||||
strength: number;
|
||||
slotIndex: number;
|
||||
strength: number;
|
||||
};
|
||||
export type MergeModelRequest = {
|
||||
voiceChangerType: VoiceChangerType;
|
||||
command: "mix";
|
||||
files: MergeElement[];
|
||||
voiceChangerType: VoiceChangerType;
|
||||
command: "mix";
|
||||
files: MergeElement[];
|
||||
};
|
||||
|
||||
@ -184,8 +184,8 @@ class EasyVC(VoiceChangerModel):
|
||||
logger.info("[Voice Changer] Pipeline is not initialized.")
|
||||
raise PipelineNotInitializedException()
|
||||
|
||||
enableTimer = True
|
||||
with Timer2("infer_easyvc", enableTimer) as t:
|
||||
enableTimer = False
|
||||
with Timer2("infer-easyvc", enableTimer) as t:
|
||||
|
||||
# 処理は16Kで実施(Pitch, embed, (infer))
|
||||
receivedData = cast(
|
||||
@ -203,7 +203,7 @@ class EasyVC(VoiceChangerModel):
|
||||
|
||||
# 入力データ生成
|
||||
data = self.generate_input(receivedData, crossfade_frame, sola_search_frame, extra_frame)
|
||||
t.record("generate_input")
|
||||
t.record("generate-input")
|
||||
|
||||
audio = data[0]
|
||||
pitchf = data[1]
|
||||
|
||||
@ -138,7 +138,7 @@ class Pipeline(object):
|
||||
# print(f"pipeline exec input, audio:{audio.shape}, pitchf:{pitchf.shape}, feature:{feature.shape}")
|
||||
# print(f"pipeline exec input, silence_front:{silence_front}, out_size:{out_size}")
|
||||
|
||||
enablePipelineTimer = True
|
||||
enablePipelineTimer = False
|
||||
with Timer2("Pipeline-Exec", enablePipelineTimer) as t: # NOQA
|
||||
# 16000のサンプリングレートで入ってきている。以降この世界は16000で処理。
|
||||
# self.t_pad = self.sr * repeat # 1秒
|
||||
|
||||
Loading…
Reference in New Issue
Block a user