Merge branch 'speakerDevice' into develop

config.py

@@ -6,7 +6,7 @@ from json import dump as json_dump
 import tkinter as tk
 from tkinter import font
 from models.translation.translation_languages import translation_lang
-from models.transcription.transcription_utils import getInputDevices, getDefaultInputDevice
+from models.transcription.transcription_utils import getInputDevices, getDefaultInputDevice, getOutputDevices, getDefaultOutputDevice
 from models.transcription.transcription_languages import transcription_lang
 from utils import generatePercentageStringsList, isUniqueStrings
 
@@ -537,6 +537,17 @@ class Config:
             self._INPUT_MIC_WORD_FILTER = sorted(set(value), key=value.index)
             saveJson(self.PATH_CONFIG, inspect.currentframe().f_code.co_name, value)
 
+    @property
+    @json_serializable('CHOICE_SPEAKER_DEVICE')
+    def CHOICE_SPEAKER_DEVICE(self):
+        return self._CHOICE_SPEAKER_DEVICE
+
+    @CHOICE_SPEAKER_DEVICE.setter
+    def CHOICE_SPEAKER_DEVICE(self, value):
+        if value in [device["name"] for device in getOutputDevices()]:
+            self._CHOICE_SPEAKER_DEVICE = value
+            saveJson(self.PATH_CONFIG, inspect.currentframe().f_code.co_name, value)
+
     @property
     @json_serializable('INPUT_SPEAKER_ENERGY_THRESHOLD')
     def INPUT_SPEAKER_ENERGY_THRESHOLD(self):
@@ -957,6 +968,7 @@ class Config:
         self._INPUT_MIC_PHRASE_TIMEOUT = 3
         self._INPUT_MIC_MAX_PHRASES = 10
         self._INPUT_MIC_WORD_FILTER = []
+        self._CHOICE_SPEAKER_DEVICE = getDefaultOutputDevice()["device"]["name"]
         self._INPUT_SPEAKER_ENERGY_THRESHOLD = 300
         self._INPUT_SPEAKER_DYNAMIC_ENERGY_THRESHOLD = False
         self._INPUT_SPEAKER_RECORD_TIMEOUT = 3

controller.py

@@ -704,6 +704,13 @@ def callbackDeleteMicWordFilter(value):
         print("There was no the target word in config.INPUT_MIC_WORD_FILTER")
 
 # Transcription (Speaker)
+def callbackSetSpeakerDevice(value):
+    print("callbackSetSpeakerDevice", value)
+    config.CHOICE_SPEAKER_DEVICE = value
+
+    model.stopCheckSpeakerEnergy()
+    view.replaceSpeakerThresholdCheckButton_Passive()
+
 def callbackSetSpeakerEnergyThreshold(value):
     print("callbackSetSpeakerEnergyThreshold", value)
     if value == "":
@@ -1046,6 +1053,8 @@ def createMainWindow(splash):
             "callback_delete_mic_word_filter": callbackDeleteMicWordFilter,
 
             # Transcription Tab (Speaker)
+            "callback_set_speaker_device": callbackSetSpeakerDevice,
+            "list_speaker_device": model.getListOutputDevice(),
             "callback_set_speaker_energy_threshold": callbackSetSpeakerEnergyThreshold,
             "callback_set_speaker_dynamic_energy_threshold": callbackSetSpeakerDynamicEnergyThreshold,
             "callback_check_speaker_threshold": callbackCheckSpeakerThreshold,

locales/en.yml

@@ -168,6 +168,9 @@ config_window:
     count_desc: "Current registered word count: %{count}"
 
 
+  speaker_device:
+    label: Speaker Device
+
   speaker_dynamic_energy_threshold:
     label_for_automatic: "Speaker Energy Threshold (Current Setting: Automatic)"
     desc_for_automatic: "Automatically determine speaker input sensitivity."

locales/ja.yml

@@ -167,6 +167,9 @@ config_window:
     count_desc: "現在登録されている単語数: %{count}"
 
 
+  speaker_device:
+    label: スピーカー （デバイス）
+
   speaker_dynamic_energy_threshold:
     label_for_automatic: "スピーカー入力感度の調整 (現在の設定: 自動)"
     desc_for_automatic: スピーカーの入力感度を自動的に調節する。

model.py

@@ -16,7 +16,7 @@ import webbrowser
 from typing import Callable
 from flashtext import KeywordProcessor
 from models.translation.translation_translator import Translator
-from models.transcription.transcription_utils import getInputDevices, getDefaultOutputDevice
+from models.transcription.transcription_utils import getInputDevices, getOutputDevices
 from models.osc.osc_tools import sendTyping, sendMessage, sendTestAction, receiveOscParameters
 from models.transcription.transcription_recorder import SelectedMicEnergyAndAudioRecorder, SelectedSpeakerEnergyAndAudioRecorder
 from models.transcription.transcription_recorder import SelectedMicEnergyRecorder, SelectedSpeakerEnergyRecorder
@@ -305,12 +305,8 @@ class Model:
         return [device["name"] for device in getInputDevices()[config.CHOICE_MIC_HOST]]
 
     @staticmethod
-    def getInputDefaultDevice():
+    def getListOutputDevice():
-        return [device["name"] for device in getInputDevices()[config.CHOICE_MIC_HOST]][0]
+        return [device["name"] for device in getOutputDevices()]
-
-    @staticmethod
-    def getOutputDefaultDevice():
-        return getDefaultOutputDevice()["name"]
 
     def startMicTranscript(self, fnc, error_fnc=None):
         if config.CHOICE_MIC_HOST == "NoHost" or config.CHOICE_MIC_DEVICE == "NoDevice":
@@ -322,14 +318,14 @@ class Model:
 
         mic_audio_queue = Queue()
         # mic_energy_queue = Queue()
-        device = [device for device in getInputDevices()[config.CHOICE_MIC_HOST] if device["name"] == config.CHOICE_MIC_DEVICE][0]
+        mic_device = [device for device in getInputDevices()[config.CHOICE_MIC_HOST] if device["name"] == config.CHOICE_MIC_DEVICE][0]
         record_timeout = config.INPUT_MIC_RECORD_TIMEOUT
         phase_timeout = config.INPUT_MIC_PHRASE_TIMEOUT
         if record_timeout > phase_timeout:
             record_timeout = phase_timeout
 
         self.mic_audio_recorder = SelectedMicEnergyAndAudioRecorder(
-            device=device,
+            device=mic_device,
             energy_threshold=config.INPUT_MIC_ENERGY_THRESHOLD,
             dynamic_energy_threshold=config.INPUT_MIC_DYNAMIC_ENERGY_THRESHOLD,
             record_timeout=record_timeout,
@@ -346,9 +342,9 @@ class Model:
             whisper_weight_type=config.WHISPER_WEIGHT_TYPE,
         )
         def sendMicTranscript():
+            try:
                 self.mic_transcriber.transcribeAudioQueue(mic_audio_queue, config.SOURCE_LANGUAGE, config.SOURCE_COUNTRY)
                 message = self.mic_transcriber.getTranscript()
-            try:
                 fnc(message)
             except Exception:
                 pass
@@ -382,7 +378,7 @@ class Model:
             self.mic_print_transcript.stop()
             self.mic_print_transcript = None
         if isinstance(self.mic_audio_recorder, SelectedMicEnergyAndAudioRecorder):
-            self.mic_audio_recorder.stop()
+            self.mic_audio_recorder.stop(wait_for_stop=False)
             self.mic_audio_recorder = None
         # if isinstance(self.mic_get_energy, threadFnc):
         #     self.mic_get_energy.stop()
@@ -418,12 +414,11 @@ class Model:
             self.mic_energy_plot_progressbar.stop()
             self.mic_energy_plot_progressbar = None
         if isinstance(self.mic_energy_recorder, SelectedMicEnergyRecorder):
-            self.mic_energy_recorder.stop()
+            self.mic_energy_recorder.stop(wait_for_stop=False)
             self.mic_energy_recorder = None
 
     def startSpeakerTranscript(self, fnc, error_fnc=None):
-        speaker_device = getDefaultOutputDevice()
+        if config.CHOICE_SPEAKER_DEVICE == "NoDevice":
-        if speaker_device["name"] == "NoDevice":
             try:
                 error_fnc()
             except Exception:
@@ -432,6 +427,7 @@ class Model:
 
         speaker_audio_queue = Queue()
         # speaker_energy_queue = Queue()
+        speaker_device = [device for device in getOutputDevices() if device["name"] == config.CHOICE_SPEAKER_DEVICE][0]
         record_timeout = config.INPUT_SPEAKER_RECORD_TIMEOUT
         phase_timeout = config.INPUT_SPEAKER_PHRASE_TIMEOUT
         if record_timeout > phase_timeout:
@@ -455,9 +451,9 @@ class Model:
             whisper_weight_type=config.WHISPER_WEIGHT_TYPE,
         )
         def sendSpeakerTranscript():
+            try:
                 self.speaker_transcriber.transcribeAudioQueue(speaker_audio_queue, config.TARGET_LANGUAGE, config.TARGET_COUNTRY)
                 message = self.speaker_transcriber.getTranscript()
-            try:
                 fnc(message)
             except Exception:
                 pass
@@ -491,15 +487,14 @@ class Model:
             self.speaker_print_transcript.stop()
             self.speaker_print_transcript = None
         if isinstance(self.speaker_audio_recorder, SelectedSpeakerEnergyAndAudioRecorder):
-            self.speaker_audio_recorder.stop()
+            self.speaker_audio_recorder.stop(wait_for_stop=False)
             self.speaker_audio_recorder = None
         # if isinstance(self.speaker_get_energy, threadFnc):
         #     self.speaker_get_energy.stop()
         #     self.speaker_get_energy = None
 
     def startCheckSpeakerEnergy(self, fnc, end_fnc, error_fnc=None):
-        speaker_device = getDefaultOutputDevice()
+        if config.CHOICE_SPEAKER_DEVICE == "NoDevice":
-        if speaker_device["name"] == "NoDevice":
             try:
                 error_fnc()
             except Exception:
@@ -516,6 +511,7 @@ class Model:
             sleep(0.01)
 
         speaker_energy_queue = Queue()
+        speaker_device = [device for device in getOutputDevices() if device["name"] == config.CHOICE_SPEAKER_DEVICE][0]
         self.speaker_energy_recorder = SelectedSpeakerEnergyRecorder(speaker_device)
         self.speaker_energy_recorder.recordIntoQueue(speaker_energy_queue)
         self.speaker_energy_plot_progressbar = threadFnc(sendSpeakerEnergy, end_fnc=end_fnc)
@@ -527,7 +523,7 @@ class Model:
             self.speaker_energy_plot_progressbar.stop()
             self.speaker_energy_plot_progressbar = None
         if isinstance(self.speaker_energy_recorder, SelectedSpeakerEnergyRecorder):
-            self.speaker_energy_recorder.stop()
+            self.speaker_energy_recorder.stop(wait_for_stop=False)
             self.speaker_energy_recorder = None
 
     def notificationXSOverlay(self, message):

models/transcription/transcription_utils.py

@@ -23,12 +23,32 @@ def getDefaultInputDevice():
 
         for host_index in range(0, p.get_host_api_count()):
             host = p.get_host_api_info_by_index(host_index)
-            for device_index in range(0, p. get_host_api_info_by_index(host_index)['deviceCount']):
+            for device_index in range(0, p.get_host_api_info_by_index(host_index)['deviceCount']):
                 device = p.get_device_info_by_host_api_device_index(host_index, device_index)
                 if device["index"] == defaultInputDevice:
                     return {"host": host, "device": device}
     return {"host": {"name": "NoHost"}, "device": {"name": "NoDevice"}}
 
+def getOutputDevices():
+    devices = []
+    with PyAudio() as p:
+        wasapi_info = p.get_host_api_info_by_type(paWASAPI)
+        for host_index in range(0, p.get_host_api_count()):
+            host = p.get_host_api_info_by_index(host_index)
+            if host["name"] == wasapi_info["name"]:
+                for device_index in range(0, p.get_host_api_info_by_index(host_index)['deviceCount']):
+                    device = p.get_device_info_by_host_api_device_index(host_index, device_index)
+                    if not device["isLoopbackDevice"]:
+                        for loopback in p.get_loopback_device_info_generator():
+                            if device["name"] in loopback["name"]:
+                                devices.append(loopback)
+
+        if len(devices) == 0:
+            devices = [{"name": "NoDevice"}]
+        else:
+            devices = [dict(t) for t in {tuple(d.items()) for d in devices}]
+    return devices
+
 def getDefaultOutputDevice():
     with PyAudio() as p:
         wasapi_info = p.get_host_api_info_by_type(paWASAPI)
@@ -42,6 +62,9 @@ def getDefaultOutputDevice():
                     if not default_speakers["isLoopbackDevice"]:
                         for loopback in p.get_loopback_device_info_generator():
                             if default_speakers["name"] in loopback["name"]:
-                                default_device = loopback
+                                return {"device": loopback}
-                                return default_device
+    return {"device": {"name": "NoDevice"}}
-    return {"name":"NoDevice"}
+
+if __name__ == "__main__":
+    print("getOutputDevices()", getOutputDevices())
+    print("getDefaultOutputDevice()", getDefaultOutputDevice())

view.py

@@ -363,6 +363,13 @@ class View():
 
 
             # Transcription Tab (Speaker)
+            VAR_LABEL_SPEAKER_DEVICE=StringVar(value=i18n.t("config_window.speaker_device.label")),
+            VAR_DESC_SPEAKER_DEVICE=None,
+            LIST_SPEAKER_DEVICE=[],
+            CALLBACK_SET_SPEAKER_DEVICE=None,
+            VAR_SPEAKER_DEVICE=StringVar(value=config.CHOICE_SPEAKER_DEVICE),
+
+
             VAR_LABEL_SPEAKER_DYNAMIC_ENERGY_THRESHOLD=StringVar(value=""),
             VAR_DESC_SPEAKER_DYNAMIC_ENERGY_THRESHOLD=StringVar(value=""),
             CALLBACK_SET_SPEAKER_DYNAMIC_ENERGY_THRESHOLD=None,
@@ -655,6 +662,9 @@ class View():
             self.view_variable.CALLBACK_DELETE_MIC_WORD_FILTER=config_window_registers.get("callback_delete_mic_word_filter", None)
 
             # Transcription Tab (Speaker)
+            self.view_variable.CALLBACK_SET_SPEAKER_DEVICE = config_window_registers.get("callback_set_speaker_device", None)
+            config_window_registers.get("list_speaker_device", None) and self.updateList_SpeakerDevice(config_window_registers["list_speaker_device"])
+
             self.view_variable.CALLBACK_SET_SPEAKER_ENERGY_THRESHOLD=config_window_registers.get("callback_set_speaker_energy_threshold", None)
             self.view_variable.CALLBACK_SET_SPEAKER_DYNAMIC_ENERGY_THRESHOLD=config_window_registers.get("callback_set_speaker_dynamic_energy_threshold", None)
             self.view_variable.CALLBACK_CHECK_SPEAKER_THRESHOLD=config_window_registers.get("callback_check_speaker_threshold", None)
@@ -721,6 +731,16 @@ class View():
             )
             self.replaceMicThresholdCheckButton_Disabled()
 
+        if config.CHOICE_SPEAKER_DEVICE == "NoDevice":
+            self.view_variable.VAR_SPEAKER_DEVICE.set("No Speaker Device Detected")
+            vrct_gui._changeConfigWindowWidgetsStatus(
+                status="disabled",
+                target_names=[
+                    "sb__optionmenu_speaker_device",
+                ]
+            )
+            self.replaceSpeakerThresholdCheckButton_Disabled()
+
         if config.USE_WHISPER_FEATURE is True:
             self.openWhisperWeightTypeWidget()
         else:
@@ -1270,6 +1290,9 @@ class View():
 
 
     def initSpeakerThresholdCheckButton(self):
+        if config.CHOICE_SPEAKER_DEVICE == "NoDevice":
+            self.replaceSpeakerThresholdCheckButton_Disabled()
+        else:
             self.replaceSpeakerThresholdCheckButton_Passive()
 
     @staticmethod
@@ -1328,6 +1351,13 @@ class View():
         vrct_gui.config_window.sb__progressbar_x_slider__progressbar_mic_energy_threshold.set(0)
 
 
+    def updateList_SpeakerDevice(self, new_speaker_device_list:list):
+        self.view_variable.LIST_SPEAKER_DEVICE = new_speaker_device_list
+        vrct_gui.dropdown_menu_window.updateDropdownMenuValues(
+            dropdown_menu_widget_id="sb__optionmenu_speaker_device",
+            dropdown_menu_values=new_speaker_device_list,
+        )
+
     def updateSetProgressBar_SpeakerEnergy(self, new_speaker_energy):
         self.updateProgressBar(
             target_progressbar_widget=vrct_gui.config_window.sb__progressbar_x_slider__progressbar_speaker_energy_threshold,

vrct_gui/_changeConfigWindowWidgetsStatus.py

@@ -43,6 +43,12 @@ def _changeConfigWindowWidgetsStatus(config_window, settings, view_variable, sta
                     disableLabelsWidgets(target_widget)
                     disableOptionmenuWidget(target_widget)
 
+            case "sb__optionmenu_speaker_device":
+                if status == "disabled":
+                    target_widget = config_window.sb__widgets["sb__optionmenu_speaker_device"]
+                    disableLabelsWidgets(target_widget)
+                    disableOptionmenuWidget(target_widget)
+
             case "sb__optionmenu_appearance_theme":
                 if status == "disabled":
                     target_widget = config_window.sb__widgets["sb__optionmenu_appearance_theme"]

vrct_gui/config_window/widgets/createSideMenuAndSettingsBoxContainers/setting_box_containers/setting_box_transcription/createSettingBox_Speaker.py

@@ -4,6 +4,7 @@ from .._SettingBoxGenerator import _SettingBoxGenerator
 
 def createSettingBox_Speaker(setting_box_wrapper, config_window, settings, view_variable):
     sbg = _SettingBoxGenerator(setting_box_wrapper, config_window, settings, view_variable)
+    createSettingBoxDropdownMenu = sbg.createSettingBoxDropdownMenu
     createSettingBoxSwitch = sbg.createSettingBoxSwitch
     createSettingBoxProgressbarXSlider = sbg.createSettingBoxProgressbarXSlider
     createSettingBoxEntry = sbg.createSettingBoxEntry
@@ -13,6 +14,9 @@ def createSettingBox_Speaker(setting_box_wrapper, config_window, settings, view_
         callFunctionIfCallable(view_variable.CALLBACK_CHECK_SPEAKER_THRESHOLD, is_turned_on)
 
 
+    def optionmenuInputSpeakerDeviceCallback(value):
+        callFunctionIfCallable(view_variable.CALLBACK_SET_SPEAKER_DEVICE, value)
+
     def sliderInputSpeakerEnergyThresholdCallback(value):
         callFunctionIfCallable(view_variable.CALLBACK_SET_SPEAKER_ENERGY_THRESHOLD, value)
 
@@ -32,6 +36,17 @@ def createSettingBox_Speaker(setting_box_wrapper, config_window, settings, view_
 
 
     row=0
+    config_window.sb__speaker_device = createSettingBoxDropdownMenu(
+        for_var_label_text=view_variable.VAR_LABEL_SPEAKER_DEVICE,
+        for_var_desc_text=view_variable.VAR_DESC_SPEAKER_DEVICE,
+        optionmenu_attr_name="sb__optionmenu_speaker_device",
+        dropdown_menu_values=view_variable.LIST_SPEAKER_DEVICE,
+        command=lambda value: optionmenuInputSpeakerDeviceCallback(value),
+        variable=view_variable.VAR_SPEAKER_DEVICE,
+    )
+    config_window.sb__speaker_device.grid(row=row)
+    row+=1
+
     config_window.sb__speaker_dynamic_energy_threshold = createSettingBoxSwitch(
         for_var_label_text=view_variable.VAR_LABEL_SPEAKER_DYNAMIC_ENERGY_THRESHOLD,
         for_var_desc_text=view_variable.VAR_DESC_SPEAKER_DYNAMIC_ENERGY_THRESHOLD,