[bugfix] Add Mic Host API

WASAPI以外のHostAPIでもマイクとして使用できるように変更
2023-07-11 22:35:42 +09:00
parent 1b8cbbd01d
commit 0b0ec88b8e
3 changed files with 48 additions and 15 deletions
--- a/VRCT.py
+++ b/VRCT.py
@@ -47,7 +47,8 @@ class App(CTk):
        self.OUTPUT_SOURCE_LANG = list(translation_lang[self.CHOICE_TRANSLATOR].keys())[1]
        self.OUTPUT_TARGET_LANG = list(translation_lang[self.CHOICE_TRANSLATOR].keys())[0]
        ## Transcription Send
-        self.CHOICE_MIC_DEVICE = get_default_input_device()["name"]
+        self.CHOICE_MIC_HOST = get_default_input_device()["host"]["name"]
+        self.CHOICE_MIC_DEVICE = get_default_input_device()["device"]["name"]
        self.INPUT_MIC_VOICE_LANGUAGE = list(transcription_lang.keys())[0]
        self.INPUT_MIC_ENERGY_THRESHOLD = 300
        self.INPUT_MIC_DYNAMIC_ENERGY_THRESHOLD = True
@@ -128,8 +129,11 @@ class App(CTk):
                    self.OUTPUT_TARGET_LANG = config["OUTPUT_TARGET_LANG"]

            # Transcription
+            if "CHOICE_MIC_HOST" in config.keys():
+                if config["CHOICE_MIC_HOST"] in [host for host in get_input_device_list().keys()]:
+                    self.CHOICE_MIC_HOST = config["CHOICE_MIC_HOST"]
            if "CHOICE_MIC_DEVICE" in config.keys():
-                if config["CHOICE_MIC_DEVICE"] in [device["name"] for device in get_input_device_list()]:
+                if config["CHOICE_MIC_DEVICE"] in [device["name"] for device in get_input_device_list()[self.CHOICE_MIC_HOST]]:
                    self.CHOICE_MIC_DEVICE = config["CHOICE_MIC_DEVICE"]
            if "INPUT_MIC_VOICE_LANGUAGE" in config.keys():
                if config["INPUT_MIC_VOICE_LANGUAGE"] in list(transcription_lang.keys()):
@@ -208,6 +212,7 @@ class App(CTk):
                "INPUT_TARGET_LANG": self.INPUT_TARGET_LANG,
                "OUTPUT_SOURCE_LANG": self.OUTPUT_SOURCE_LANG,
                "OUTPUT_TARGET_LANG": self.OUTPUT_TARGET_LANG,
+                "CHOICE_MIC_HOST": self.CHOICE_MIC_HOST,
                "CHOICE_MIC_DEVICE": self.CHOICE_MIC_DEVICE,
                "INPUT_MIC_VOICE_LANGUAGE": self.INPUT_MIC_VOICE_LANGUAGE,
                "INPUT_MIC_ENERGY_THRESHOLD": self.INPUT_MIC_ENERGY_THRESHOLD,
@@ -459,7 +464,7 @@ class App(CTk):
        if self.ENABLE_TRANSCRIPTION_SEND is True:
            self.button_config.configure(state="disabled", fg_color=["gray92", "gray14"])
            self.mic_audio_queue = Queue()
-            mic_device = [device for device in get_input_device_list() if device["name"] == self.CHOICE_MIC_DEVICE][0]
+            mic_device = [device for device in get_input_device_list()[self.CHOICE_MIC_HOST] if device["name"] == self.CHOICE_MIC_DEVICE][0]
            self.mic_audio_recorder = SelectedMicRecorder(
                mic_device,
                self.INPUT_MIC_ENERGY_THRESHOLD,
--- a/audio_utils.py
+++ b/audio_utils.py
@@ -1,14 +1,17 @@
 from pyaudiowpatch import PyAudio, paWASAPI

 def get_input_device_list():
-    devices = []
+    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()):
-            for device_index in range(0, p. get_host_api_info_by_index(host_index)['deviceCount']):
+            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']):
                device = p.get_device_info_by_host_api_device_index(host_index, device_index)
-                if device["hostApi"] == wasapi_info["index"] and device["maxInputChannels"] > 0 and device["isLoopbackDevice"] is False:
-                    devices.append(device)
+                if device["maxInputChannels"] > 0 and device["isLoopbackDevice"] is False:
+                    if host["name"] in devices.keys():
+                        devices[host["name"]].append(device)
+                    else:
+                        devices[host["name"]] = [device]
    return devices

 def get_output_device_list():
@@ -22,15 +25,15 @@ def get_output_device_list():

 def get_default_input_device():
    with PyAudio() as p:
-        wasapi_info = p.get_host_api_info_by_type(paWASAPI)
-        defaultInputDevice = wasapi_info["defaultInputDevice"]
+        api_info = p.get_default_host_api_info()
+        defaultInputDevice = api_info["defaultInputDevice"]

        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']):
                device = p.get_device_info_by_host_api_device_index(host_index, device_index)
                if device["index"] == defaultInputDevice:
-                    default_device = device
-                    return default_device
+                    return {"host":host, "device": device}

 def get_default_output_device():
    with PyAudio() as p:
--- a/window_config.py
+++ b/window_config.py
@@ -1,6 +1,5 @@
 from time import sleep
 from queue import Queue
-from threading import Thread
 from os import path as os_path
 from tkinter import DoubleVar, IntVar 
 from tkinter import font as tk_font
@@ -239,6 +238,24 @@ class ToplevelWindowConfig(CTkToplevel):
        row = 0
        padx = 5
        pady = 1
+        self.label_input_mic_host = CTkLabel(
+            self.tabview_config.tab("Transcription"),
+            text="Input Mic Host:",
+            fg_color="transparent",
+            font=CTkFont(family=self.parent.FONT_FAMILY)
+        )
+        self.label_input_mic_host.grid(row=row, column=0, columnspan=1, padx=padx, pady=pady, sticky="nsw")
+        self.optionmenu_input_mic_host = CTkOptionMenu(
+            self.tabview_config.tab("Transcription"),
+            values=[host for host in get_input_device_list().keys()],
+            command=self.optionmenu_input_mic_host_callback,
+            variable=StringVar(value=self.parent.CHOICE_MIC_HOST),
+            font=CTkFont(family=self.parent.FONT_FAMILY),
+        )
+        self.optionmenu_input_mic_host.grid(row=row, column=1, columnspan=1 ,padx=padx, pady=pady, sticky="nsew")
+        self.optionmenu_input_mic_host._dropdown_menu.configure(font=CTkFont(family=self.parent.FONT_FAMILY))
+
+        row += 1
        self.label_input_mic_device = CTkLabel(
            self.tabview_config.tab("Transcription"),
            text="Input Mic Device:",
@@ -248,7 +265,7 @@ class ToplevelWindowConfig(CTkToplevel):
        self.label_input_mic_device.grid(row=row, column=0, columnspan=1, padx=padx, pady=pady, sticky="nsw")
        self.optionmenu_input_mic_device = CTkOptionMenu(
            self.tabview_config.tab("Transcription"),
-            values=[device["name"] for device in get_input_device_list()],
+            values=[device["name"] for device in get_input_device_list()[self.parent.CHOICE_MIC_HOST]],
            command=self.optionmenu_input_mic_device_callback,
            variable=StringVar(value=self.parent.CHOICE_MIC_DEVICE),
            font=CTkFont(family=self.parent.FONT_FAMILY),
@@ -710,6 +727,9 @@ class ToplevelWindowConfig(CTkToplevel):
        self.optionmenu_translation_output_target_language._dropdown_menu.configure(font=CTkFont(family=choice))

        # tab Transcription
+        self.label_input_mic_host.configure(font=CTkFont(family=choice))
+        self.optionmenu_input_mic_host.configure(font=CTkFont(family=choice))
+        self.optionmenu_input_mic_host._dropdown_menu.configure(font=CTkFont(family=choice))
        self.label_input_mic_device.configure(font=CTkFont(family=choice))
        self.optionmenu_input_mic_device.configure(font=CTkFont(family=choice))
        self.optionmenu_input_mic_device._dropdown_menu.configure(font=CTkFont(family=choice))
@@ -822,6 +842,11 @@ class ToplevelWindowConfig(CTkToplevel):
        self.parent.OUTPUT_TARGET_LANG = choice
        save_json(self.parent.PATH_CONFIG, "OUTPUT_TARGET_LANG", self.parent.OUTPUT_TARGET_LANG)

+    def optionmenu_input_mic_host_callback(self, choice):
+        self.parent.CHOICE_MIC_HOST = choice
+        save_json(self.parent.PATH_CONFIG, "CHOICE_MIC_HOST", self.parent.CHOICE_MIC_HOST)
+        self.optionmenu_input_mic_device.configure(values=[device["name"] for device in get_input_device_list()[self.parent.CHOICE_MIC_HOST]])
+
    def optionmenu_input_mic_device_callback(self, choice):
        self.parent.CHOICE_MIC_DEVICE = choice
        save_json(self.parent.PATH_CONFIG, "CHOICE_MIC_DEVICE", self.parent.CHOICE_MIC_DEVICE)
@@ -844,7 +869,7 @@ class ToplevelWindowConfig(CTkToplevel):
    def checkbox_input_mic_threshold_check_callback(self):
        if self.checkbox_input_mic_threshold_check.get():
            self.mic_energy_queue = Queue()
-            mic_device = [device for device in get_input_device_list() if device["name"] == self.parent.CHOICE_MIC_DEVICE][0]
+            mic_device = [device for device in get_input_device_list()[self.parent.CHOICE_MIC_HOST] if device["name"] == self.parent.CHOICE_MIC_DEVICE][0]
            self.mic_energy_recorder = SelectedMicEnergyRecorder(mic_device)
            self.mic_energy_recorder.record_into_queue(self.mic_energy_queue)
            self.mic_energy_plot_progressbar = thread_fnc(self.progressBar_input_mic_energy_plot)