separate server device fin

server/sio/MMVC_Namespace.py

@@ -24,7 +24,8 @@ class MMVC_Namespace(socketio.AsyncNamespace):
     def __init__(self, namespace: str, voiceChangerManager: VoiceChangerManager):
         super().__init__(namespace)
         self.voiceChangerManager = voiceChangerManager
-        self.voiceChangerManager.voiceChanger.emitTo = self.emit_coroutine
+        # self.voiceChangerManager.voiceChanger.emitTo = self.emit_coroutine
+        self.voiceChangerManager.setEmitTo(self.emit_coroutine)
 
     @classmethod
     def get_instance(cls, voiceChangerManager: VoiceChangerManager):
@@ -34,11 +35,7 @@ class MMVC_Namespace(socketio.AsyncNamespace):
 
     def on_connect(self, sid, environ):
         self.sid = sid
-        print(
+        print("[{}] connet sid : {}".format(datetime.now().strftime("%Y-%m-%d %H:%M:%S"), sid))
-            "[{}] connet sid : {}".format(
-                datetime.now().strftime("%Y-%m-%d %H:%M:%S"), sid
-            )
-        )
         pass
 
     async def on_request_message(self, sid, msg):
@@ -50,9 +47,7 @@ class MMVC_Namespace(socketio.AsyncNamespace):
             print(data)
             await self.emit("response", [timestamp, 0], to=sid)
         else:
-            unpackedData = np.array(
+            unpackedData = np.array(struct.unpack("<%sh" % (len(data) // struct.calcsize("<h")), data)).astype(np.int16)
-                struct.unpack("<%sh" % (len(data) // struct.calcsize("<h")), data)
-            ).astype(np.int16)
 
             res = self.voiceChangerManager.changeVoice(unpackedData)
             audio1 = res[0]

server/voice_changer/Local/ServerDevice.py

@@ -107,20 +107,16 @@ class ServerDevice:
                 sd._initialize()
 
                 sd.default.device[0] = self.settings.serverInputDeviceId
-                # currentInputDeviceId = self.settings.serverInputDeviceId
                 sd.default.device[1] = self.settings.serverOutputDeviceId
-                # currentOutputDeviceId = self.settings.serverOutputDeviceId
 
                 serverInputAudioDevice = self.getServerInputAudioDevice(sd.default.device[0])
                 serverOutputAudioDevice = self.getServerOutputAudioDevice(sd.default.device[1])
-                print(serverInputAudioDevice, serverOutputAudioDevice)
+                print("Devices:", serverInputAudioDevice, serverOutputAudioDevice)
                 if serverInputAudioDevice is None or serverOutputAudioDevice is None:
                     time.sleep(2)
                     print("serverInputAudioDevice or serverOutputAudioDevice is None")
                     continue
 
-                # currentInputChannelNum = serverInputAudioDevice.maxInputChannels
-                # currentOutputChannelNum = serverOutputAudioDevice.maxOutputChannels
                 sd.default.channels[0] = serverInputAudioDevice.maxInputChannels
                 sd.default.channels[1] = serverOutputAudioDevice.maxOutputChannels
 
@@ -164,10 +160,7 @@ class ServerDevice:
                     ):
                         while self.settings.serverAudioStated == 1 and sd.default.device[0] == self.settings.serverInputDeviceId and sd.default.device[1] == self.settings.serverOutputDeviceId and currentModelSamplingRate == self.serverDeviceCallbacks.get_processing_sampling_rate() and currentInputChunkNum == self.settings.serverReadChunkSize:
                             time.sleep(2)
-                            print(
+                            print("[Voice Changer] server audio", self.performance)
-                                "[Voice Changer] server audio",
-                                self.performance,
-                            )
                             print(f"[Voice Changer] started:{self.settings.serverAudioStated}, input:{sd.default.device[0]}, output:{sd.default.device[1]}, mic_sr:{self.settings.serverInputAudioSampleRate}, model_sr:{currentModelSamplingRate}, chunk:{currentInputChunkNum}, ch:[{sd.default.channels}]")
 
                 except Exception as e:

server/voice_changer/RVC/RVC.py

@@ -340,7 +340,6 @@ class RVC:
 
         audio = torchaudio.functional.resample(audio, self.settings.modelSamplingRate, 16000, rolloff=0.99)
         repeat = 1 if self.settings.rvcQuality else 0
-        print()
         sid = 0
         f0_up_key = self.settings.tran
         index_rate = self.settings.indexRatio

server/voice_changer/VoiceChangerManager.py

@@ -8,6 +8,8 @@ from voice_changer.utils.VoiceChangerParams import VoiceChangerParams
 from dataclasses import dataclass, asdict
 import torch
 import threading
+from typing import Callable
+from typing import Any
 
 
 @dataclass()
@@ -34,7 +36,7 @@ class VoiceChangerManager(ServerDeviceCallbacks):
         return self.changeVoice(unpackedData)
 
     def emitTo(self, performance: list[float]):
-        print("emit ", performance)
+        self.emitToFunc(performance)
 
     def get_processing_sampling_rate(self):
         return self.voiceChanger.get_processing_sampling_rate()
@@ -139,3 +141,6 @@ class VoiceChangerManager(ServerDeviceCallbacks):
 
     def upload_model_assets(self, params: str):
         return self.voiceChanger.upload_model_assets(params)
+
+    def setEmitTo(self, emitTo: Callable[[Any], None]):
+        self.emitToFunc = emitTo