Merge branch 'change_device_with_energy' into for_webui

src-python/model.py

@@ -69,6 +69,7 @@ class Model:
 
     def init(self):
         self.logger = None
+        self.th_check_device = None
         self.mic_print_transcript = None
         self.mic_audio_recorder = None
         self.mic_energy_recorder = None
@@ -422,6 +423,33 @@ class Model:
     def getListOutputDevice():
         return [device["name"] for device in getOutputDevices()]
 
+    def startAutomaticDeviceSelection(self, fnc_mic, fnc_speaker):
+        def checkDevice(fnc_mic, fnc_speaker):
+            if config.ENABLE_MIC_AUTOMATIC_SELECTION is True:
+                default_device = getDefaultInputDevice()
+                mic_host_name = default_device["host"]["name"]
+                mic_device_name = default_device["device"]["name"]
+                if mic_host_name != config.CHOICE_MIC_HOST or mic_device_name != config.CHOICE_MIC_DEVICE:
+                    fnc_mic(mic_host_name, mic_device_name)
+
+            if config.ENABLE_SPEAKER_AUTOMATIC_SELECTION is True:
+                default_device = getDefaultOutputDevice()
+                speaker_device_name = default_device["device"]["name"]
+                if speaker_device_name != config.CHOICE_SPEAKER_DEVICE:
+                    fnc_speaker(speaker_device_name)
+            sleep(1)
+
+        if isinstance(self.th_check_device, threadFnc) is False:
+            self.th_check_device = threadFnc(checkDevice, args=(fnc_mic, fnc_speaker,))
+            self.th_check_device.daemon = True
+            self.th_check_device.start()
+
+    def stopAutomaticDeviceSelection(self):
+        if config.ENABLE_MIC_AUTOMATIC_SELECTION is False and config.ENABLE_SPEAKER_AUTOMATIC_SELECTION is False:
+            if isinstance(self.th_check_device, threadFnc):
+                self.th_check_device.stop()
+                self.th_check_device = None
+
     def startMicTranscript(self, fnc):
         if config.ENABLE_MIC_AUTOMATIC_SELECTION is True:
             default_device = getDefaultInputDevice()
@@ -553,7 +581,10 @@ class Model:
         #     self.mic_get_energy.stop()
         #     self.mic_get_energy = None
 
-    def startCheckMicEnergy(self, fnc):
+    def startCheckMicEnergy(self, fnc:Callable[[float], None]=None) -> None:
+        if isinstance(fnc, Callable):
+            self.check_mic_energy_fnc = fnc
+
         if config.ENABLE_MIC_AUTOMATIC_SELECTION is True:
             default_device = getDefaultInputDevice()
             mic_host_name = default_device["host"]["name"]
@@ -572,7 +603,7 @@ class Model:
             if mic_energy_queue.empty() is False:
                 energy = mic_energy_queue.get()
                 try:
-                    fnc(energy)
+                    self.check_mic_energy_fnc(energy)
                 except Exception:
                     pass
             sleep(0.01)
@@ -684,7 +715,10 @@ class Model:
         #     self.speaker_get_energy.stop()
         #     self.speaker_get_energy = None
 
-    def startCheckSpeakerEnergy(self, fnc):
+    def startCheckSpeakerEnergy(self, fnc:Callable[[float], None]=None) -> None:
+        if isinstance(fnc, Callable):
+            self.check_speaker_energy_fnc = fnc
+
         if config.ENABLE_SPEAKER_AUTOMATIC_SELECTION is True:
             default_device = getDefaultOutputDevice()
             speaker_device_name = default_device["device"]["name"]
@@ -701,7 +735,7 @@ class Model:
             if speaker_energy_queue.empty() is False:
                 energy = speaker_energy_queue.get()
                 try:
-                    fnc(energy)
+                    self.check_speaker_energy_fnc(energy)
                 except Exception:
                     pass
             sleep(0.01)

src-python/webui_controller.py

@@ -5,7 +5,7 @@ from threading import Thread
 import re
 from config import config
 from model import model
-from utils import isUniqueStrings, printLog, printResponse
+from utils import isUniqueStrings, printLog
 
 # Common
 class DownloadSoftwareProgressBar:
@@ -778,22 +778,60 @@ def callbackClearDeeplAuthKey(*args, **kwargs) -> dict:
 
 # Transcription Tab
 # Transcription (Mic)
+class UpdateSelectedDevice:
+    def __init__(self, action):
+        self.action = action
 
-def callbackEnableMicAutomaticSelection(*args, **kwargs) -> dict:
+    def set_mic(self, host, device) -> None:
+        config.CHOICE_MIC_HOST = host
+        config.CHOICE_MIC_DEVICE = device
+        printLog("Update Host/Mic Device", f"{host}/{device}")
+        self.action("mic", {
+            "status":200,
+            "result":{"host":host, "device":device}
+            })
+
+    def set_speaker(self, device) -> None:
+        config.CHOICE_SPEAKER_DEVICE = device
+        printLog("Update Speaker Device", device)
+        self.action("speaker", {
+            "status":200,
+            "result":device
+            })
+
+def callbackEnableMicAutomaticSelection(data, action, *args, **kwargs) -> dict:
     printLog("Enable Mic Automatic Selection")
+    update_device = UpdateSelectedDevice(action)
+    model.startAutomaticDeviceSelection(update_device.set_mic, update_device.set_speaker)
     config.ENABLE_MIC_AUTOMATIC_SELECTION = True
     return {"status":200, "result":config.ENABLE_MIC_AUTOMATIC_SELECTION}
 
 def callbackDisableMicAutomaticSelection(*args, **kwargs) -> dict:
     printLog("Disable Mic Automatic Selection")
+    model.stopAutomaticDeviceSelection()
     config.ENABLE_MIC_AUTOMATIC_SELECTION = False
     return {"status":200, "result":config.ENABLE_MIC_AUTOMATIC_SELECTION}
 
+def callbackEnableSpeakerAutomaticSelection(data, action, *args, **kwargs) -> dict:
+    printLog("Enable Speaker Automatic Selection")
+    update_device = UpdateSelectedDevice(action)
+    model.startAutomaticDeviceSelection(update_device.set_mic, update_device.set_speaker)
+    config.ENABLE_SPEAKER_AUTOMATIC_SELECTION = True
+    return {"status":200, "result":config.ENABLE_SPEAKER_AUTOMATIC_SELECTION}
+
+def callbackDisableSpeakerAutomaticSelection(*args, **kwargs) -> dict:
+    printLog("Disable Speaker Automatic Selection")
+    model.stopAutomaticDeviceSelection()
+    config.ENABLE_SPEAKER_AUTOMATIC_SELECTION = False
+    return {"status":200, "result":config.ENABLE_SPEAKER_AUTOMATIC_SELECTION}
+
 def callbackSetMicHost(data, *args, **kwargs) -> dict:
     printLog("Set Mic Host", data)
     config.CHOICE_MIC_HOST = data
     config.CHOICE_MIC_DEVICE = model.getInputDefaultDevice()
-    model.stopCheckMicEnergy()
+    if config.ENABLE_CHECK_ENERGY_SEND is True:
+        model.stopCheckMicEnergy()
+        model.startCheckMicEnergy()
     return {"status":200,
             "result":{
                 "host":config.CHOICE_MIC_HOST,
@@ -804,8 +842,10 @@ def callbackSetMicHost(data, *args, **kwargs) -> dict:
 def callbackSetMicDevice(data, *args, **kwargs) -> dict:
     printLog("Set Mic Device", data)
     config.CHOICE_MIC_DEVICE = data
-    model.stopCheckMicEnergy()
+    if config.ENABLE_CHECK_ENERGY_SEND is True:
-    return {"status":200, "result":config.CHOICE_MIC_DEVICE}
+        model.stopCheckMicEnergy()
+        model.startCheckMicEnergy()
+    return {"status":200, "result":{"host":config.CHOICE_MIC_HOST, "device":config.CHOICE_MIC_DEVICE}}
 
 def callbackSetMicEnergyThreshold(data, *args, **kwargs) -> dict:
     printLog("Set Mic Energy Threshold", data)
@@ -926,21 +966,12 @@ def callbackDeleteMicWordFilter(data, *args, **kwargs) -> dict:
     return {"status":200, "result":config.INPUT_MIC_WORD_FILTER}
 
 # Transcription (Speaker)
-
-def callbackEnableSpeakerAutomaticSelection(*args, **kwargs) -> dict:
-    printLog("Enable Speaker Automatic Selection")
-    config.ENABLE_SPEAKER_AUTOMATIC_SELECTION = True
-    return {"status":200, "result":config.ENABLE_SPEAKER_AUTOMATIC_SELECTION}
-
-def callbackDisableSpeakerAutomaticSelection(*args, **kwargs) -> dict:
-    printLog("Disable Speaker Automatic Selection")
-    config.ENABLE_SPEAKER_AUTOMATIC_SELECTION = False
-    return {"status":200, "result":config.ENABLE_SPEAKER_AUTOMATIC_SELECTION}
-
 def callbackSetSpeakerDevice(data, *args, **kwargs) -> dict:
     printLog("Set Speaker Device", data)
     config.CHOICE_SPEAKER_DEVICE = data
-    model.stopCheckSpeakerEnergy()
+    if config.ENABLE_CHECK_ENERGY_RECEIVE is True:
+        model.stopCheckSpeakerEnergy()
+        model.startCheckSpeakerEnergy()
     return {"status":200, "result":config.CHOICE_SPEAKER_DEVICE}
 
 def callbackSetSpeakerEnergyThreshold(data, *args, **kwargs) -> dict:
@@ -1325,7 +1356,7 @@ def getListInputDevice(*args, **kwargs) -> dict:
 def getListOutputDevice(*args, **kwargs) -> dict:
     return {"status":200, "result": model.getListOutputDevice()}
 
-def init(endpoints:dict, *args, **kwargs) -> None:
+def init(actions:dict, *args, **kwargs) -> None:
     printLog("Start Initialization")
 
     printLog("Start check DeepL API Key")
@@ -1343,9 +1374,8 @@ def init(endpoints:dict, *args, **kwargs) -> None:
     # check Downloaded CTranslate2 Model Weight
     printLog("Check Downloaded CTranslate2 Model Weight")
     if config.USE_TRANSLATION_FEATURE is True and model.checkCTranslatorCTranslate2ModelWeight() is False:
-        def callback(progress):
+        download = DownloadCTranslate2ProgressBar(actions["download_ctranslate2"])
-            printResponse(200, endpoints["ctranslate2"], {"progress":progress})
+        startThreadingDownloadCtranslate2Weight(download.set)
-        startThreadingDownloadCtranslate2Weight(callback)
 
     # set Transcription Engine
     printLog("Set Transcription Engine")
@@ -1357,9 +1387,8 @@ def init(endpoints:dict, *args, **kwargs) -> None:
     # check Downloaded Whisper Model Weight
     printLog("Check Downloaded Whisper Model Weight")
     if config.USE_WHISPER_FEATURE is True and model.checkTranscriptionWhisperModelWeight() is False:
-        def callback(progress):
+        download = DownloadWhisperProgressBar(actions["download_whisper"])
-            printResponse(200, endpoints["whisper"], {"progress":progress})
+        startThreadingDownloadWhisperWeight(download.set)
-        startThreadingDownloadWhisperWeight(callback)
 
     # set word filter
     printLog("Set Word Filter")
@@ -1380,4 +1409,10 @@ def init(endpoints:dict, *args, **kwargs) -> None:
     model.startReceiveOSC()
     if config.ENABLE_VRC_MIC_MUTE_SYNC is True:
         model.startCheckMuteSelfStatus()
-    printLog("End Initialization")
+    printLog("End Initialization")
+
+    # init Auto device selection
+    printLog("Init Auto Device Selection")
+    if config.ENABLE_MIC_AUTOMATIC_SELECTION is True or config.ENABLE_SPEAKER_AUTOMATIC_SELECTION is True:
+        update_device = UpdateSelectedDevice(actions["update_selected_device"])
+        model.startAutomaticDeviceSelection(update_device.set_mic, update_device.set_speaker)

src-python/webui_mainloop.py

@@ -240,6 +240,14 @@ action_mapping = {
     "/controller/callback_download_whisper_weight": {
         "download":"/action/download_whisper_weight"
     },
+    "/controller/callback_enable_mic_automatic_selection": {
+        "mic":"/controller/callback_set_mic_host",
+        "speaker":"/controller/callback_set_speaker_device",
+    },
+    "/controller/callback_enable_speaker_automatic_selection": {
+        "mic":"/controller/callback_set_mic_host",
+        "speaker":"/controller/callback_set_speaker_device",
+    }
 }
 
 def handleConfigRequest(endpoint):
@@ -259,12 +267,16 @@ def handleControllerRequest(endpoint, data=None):
         status = 404
     else:
         action_endpoint = action_mapping.get(endpoint, None)
-        if action_endpoint is not None:
+        try:
-            response = handler(data, Action(action_endpoint).transmit)
+            if action_endpoint is not None:
-        else:
+                response = handler(data, Action(action_endpoint).transmit)
-            response = handler(data)
+            else:
-        status = response.get("status", None)
+                response = handler(data)
-        result = response.get("result", None)
+            status = response.get("status", None)
+            result = response.get("result", None)
+        except Exception as e:
+            result = str(e)
+            status = 500
     return result, status
 
 class Action:
@@ -272,9 +284,12 @@ class Action:
         self.endpoints = endpoints
 
     def transmit(self, key:str, data:dict) -> None:
-        status = data.get("status", None)
+        if key not in self.endpoints:
-        result = data.get("result", None)
+            printLog("Invalid endpoint", key)
-        printResponse(status, self.endpoints[key], result)
+        else:
+            status = data.get("status", None)
+            result = data.get("result", None)
+            printResponse(status, self.endpoints[key], result)
 
 def main():
     received_data = sys.stdin.readline().strip()
@@ -284,7 +299,7 @@ def main():
         endpoint = received_data.get("endpoint", None)
         data = received_data.get("data", None)
         data = encodeBase64(data) if data is not None else None
-        printLog(endpoint, data)
+        printLog(endpoint, {"receive_data":data})
 
         try:
             match endpoint.split("/")[1]:
@@ -297,12 +312,14 @@ def main():
         except Exception as e:
             result = str(e)
             status = 500
+        printLog(endpoint, {"send_data":result})
         printResponse(status, endpoint, result)
 
 if __name__ == "__main__":
     controller.init({
-        "ctranslate2": action_mapping["/controller/callback_download_ctranslate2_weight"]["download"],
+        "download_ctranslate2": Action(action_mapping["/controller/callback_download_ctranslate2_weight"]).transmit,
-        "whisper": action_mapping["/controller/callback_download_whisper_weight"]["download"],
+        "download_whisper": Action(action_mapping["/controller/callback_download_whisper_weight"]).transmit,
+        "update_selected_device": Action(action_mapping["/controller/callback_enable_mic_automatic_selection"]).transmit,
     })
 
     process = "main"
@@ -317,12 +334,11 @@ if __name__ == "__main__":
                     traceback.print_exc(file=f)
 
         case "test":
-            endpoint = "/controller/callback_download_ctranslate2_weight"
+            for _ in range(100):
-            result, status = handleControllerRequest(endpoint)
+                time.sleep(0.5)
-            printResponse(status, endpoint, result)
+                endpoint = "/controller/list_mic_host"
-            endpoint = "/controller/callback_download_whisper_weight"
+                result, status = handleControllerRequest(endpoint)
-            result, status = handleControllerRequest(endpoint)
+                printResponse(status, endpoint, result)
-            printResponse(status, endpoint, result)
 
         case "test_all":
             import time