every_holiday/VRCT
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

update transcription UI

Browse Source
This commit is contained in:
misyaguziya
2023-06-13 18:04:29 +09:00
parent 61f769ef06
commit dbc322ac15
3 changed files with 189 additions and 162 deletions
Download Patch File Download Diff File Expand all files Collapse all files

157
VRCT.py
View File

@@ -35,12 +35,14 @@ class App(customtkinter.CTk):
## Transcription ## Transcription
self.CHOICE_MIC_DEVICE = None self.CHOICE_MIC_DEVICE = None
self.INPUT_MIC_VOICE_LANGUAGE = "ja-JP" self.INPUT_MIC_VOICE_LANGUAGE = "ja-JP"
self.ENABLE_MIC_IS_DYNAMIC = False self.INPUT_MIC_IS_DYNAMIC = False
self.MIC_THRESHOLD = 300 self.INPUT_MIC_THRESHOLD = 300
self.CHOICE_SPEAKER_DEVICE = None self.CHOICE_SPEAKER_DEVICE = None
self.INPUT_SPEAKER_VOICE_LANGUAGE = "ja-JP" self.INPUT_SPEAKER_VOICE_LANGUAGE = "ja-JP"
self.ENABLE_SPEAKER_IS_DYNAMIC = False self.INPUT_SPEAKER_SAMPLING_RATE = 16000
self.SPEAKER_THRESHOLD = 300 self.INPUT_SPEAKER_INTERVAL = 3
self.INPUT_SPEAKER_BUFFER_SIZE = 4096
## Parameter ## Parameter
self.OSC_IP_ADDRESS = "127.0.0.1" self.OSC_IP_ADDRESS = "127.0.0.1"
self.OSC_PORT = 9000 self.OSC_PORT = 9000
@@ -91,18 +93,20 @@ class App(customtkinter.CTk):
self.CHOICE_MIC_DEVICE = config["CHOICE_MIC_DEVICE"] self.CHOICE_MIC_DEVICE = config["CHOICE_MIC_DEVICE"]
if "INPUT_MIC_VOICE_LANGUAGE" in config.keys(): if "INPUT_MIC_VOICE_LANGUAGE" in config.keys():
self.INPUT_MIC_VOICE_LANGUAGE = config["INPUT_MIC_VOICE_LANGUAGE"] self.INPUT_MIC_VOICE_LANGUAGE = config["INPUT_MIC_VOICE_LANGUAGE"]
if "ENABLE_MIC_IS_DYNAMIC" in config.keys(): if "INPUT_MIC_IS_DYNAMIC" in config.keys():
self.ENABLE_MIC_IS_DYNAMIC = config["ENABLE_MIC_IS_DYNAMIC"] self.INPUT_MIC_IS_DYNAMIC = config["INPUT_MIC_IS_DYNAMIC"]
if "MIC_THRESHOLD" in config.keys(): if "INPUT_MIC_THRESHOLD" in config.keys():
self.MIC_THRESHOLD = config["MIC_THRESHOLD"] self.INPUT_MIC_THRESHOLD = config["INPUT_MIC_THRESHOLD"]
if "CHOICE_SPEAKER_DEVICE" in config.keys(): if "CHOICE_SPEAKER_DEVICE" in config.keys():
self.CHOICE_SPEAKER_DEVICE = config["CHOICE_SPEAKER_DEVICE"] self.CHOICE_SPEAKER_DEVICE = config["CHOICE_SPEAKER_DEVICE"]
if "INPUT_SPEAKER_VOICE_LANGUAGE" in config.keys(): if "INPUT_SPEAKER_VOICE_LANGUAGE" in config.keys():
self.INPUT_SPEAKER_VOICE_LANGUAGE = config["INPUT_SPEAKER_VOICE_LANGUAGE"] self.INPUT_SPEAKER_VOICE_LANGUAGE = config["INPUT_SPEAKER_VOICE_LANGUAGE"]
if "ENABLE_SPEAKER_IS_DYNAMIC" in config.keys(): if "INPUT_SPEAKER_SAMPLING_RATE" in config.keys():
self.ENABLE_SPEAKER_IS_DYNAMIC = config["ENABLE_SPEAKER_IS_DYNAMIC"] self.INPUT_SPEAKER_SAMPLING_RATE = config["INPUT_SPEAKER_SAMPLING_RATE"]
if "SPEAKER_THRESHOLD" in config.keys(): if "INPUT_SPEAKER_INTERVAL" in config.keys():
self.SPEAKER_THRESHOLD = config["SPEAKER_THRESHOLD"] self.INPUT_SPEAKER_INTERVAL = config["INPUT_SPEAKER_INTERVAL"]
if "INPUT_SPEAKER_BUFFER_SIZE" in config.keys():
self.INPUT_SPEAKER_BUFFER_SIZE = config["INPUT_SPEAKER_BUFFER_SIZE"]
# Parameter # Parameter
if "OSC_IP_ADDRESS" in config.keys(): if "OSC_IP_ADDRESS" in config.keys():
@@ -130,12 +134,13 @@ class App(customtkinter.CTk):
"OUTPUT_TARGET_LANG": self.OUTPUT_TARGET_LANG, "OUTPUT_TARGET_LANG": self.OUTPUT_TARGET_LANG,
"CHOICE_MIC_DEVICE": self.CHOICE_MIC_DEVICE, "CHOICE_MIC_DEVICE": self.CHOICE_MIC_DEVICE,
"INPUT_MIC_VOICE_LANGUAGE": self.INPUT_MIC_VOICE_LANGUAGE, "INPUT_MIC_VOICE_LANGUAGE": self.INPUT_MIC_VOICE_LANGUAGE,
"ENABLE_MIC_IS_DYNAMIC": self.ENABLE_MIC_IS_DYNAMIC, "INPUT_MIC_IS_DYNAMIC": self.INPUT_MIC_IS_DYNAMIC,
"MIC_THRESHOLD": self.MIC_THRESHOLD, "INPUT_MIC_THRESHOLD": self.INPUT_MIC_THRESHOLD,
"CHOICE_SPEAKER_DEVICE": self.CHOICE_SPEAKER_DEVICE, "CHOICE_SPEAKER_DEVICE": self.CHOICE_SPEAKER_DEVICE,
"INPUT_SPEAKER_VOICE_LANGUAGE": self.INPUT_SPEAKER_VOICE_LANGUAGE, "INPUT_SPEAKER_VOICE_LANGUAGE": self.INPUT_SPEAKER_VOICE_LANGUAGE,
"ENABLE_SPEAKER_IS_DYNAMIC": self.ENABLE_SPEAKER_IS_DYNAMIC, "INPUT_SPEAKER_SAMPLING_RATE": self.INPUT_SPEAKER_SAMPLING_RATE,
"SPEAKER_THRESHOLD": self.SPEAKER_THRESHOLD, "INPUT_SPEAKER_INTERVAL": self.INPUT_SPEAKER_INTERVAL,
"INPUT_SPEAKER_BUFFER_SIZE": self.INPUT_SPEAKER_BUFFER_SIZE,
"OSC_IP_ADDRESS": self.OSC_IP_ADDRESS, "OSC_IP_ADDRESS": self.OSC_IP_ADDRESS,
"OSC_PORT": self.OSC_PORT, "OSC_PORT": self.OSC_PORT,
"AUTH_KEYS": self.AUTH_KEYS, "AUTH_KEYS": self.AUTH_KEYS,
@@ -215,29 +220,40 @@ class App(customtkinter.CTk):
self.tabview_logs = customtkinter.CTkTabview(master=self) self.tabview_logs = customtkinter.CTkTabview(master=self)
self.tabview_logs.add("send") self.tabview_logs.add("send")
self.tabview_logs.add("receive") self.tabview_logs.add("receive")
self.tabview_logs.add("system")
self.tabview_logs.grid(row=0, column=1, padx=5, pady=0, sticky="nsew") self.tabview_logs.grid(row=0, column=1, padx=5, pady=0, sticky="nsew")
self.tabview_logs._segmented_button.grid(sticky="W") self.tabview_logs._segmented_button.grid(sticky="W")
self.tabview_logs.tab("send").grid_rowconfigure(0, weight=1) self.tabview_logs.tab("send").grid_rowconfigure(0, weight=1)
self.tabview_logs.tab("send").grid_columnconfigure(0, weight=1) self.tabview_logs.tab("send").grid_columnconfigure(0, weight=1)
self.tabview_logs.tab("receive").grid_rowconfigure(0, weight=1) self.tabview_logs.tab("receive").grid_rowconfigure(0, weight=1)
self.tabview_logs.tab("receive").grid_columnconfigure(0, weight=1) self.tabview_logs.tab("receive").grid_columnconfigure(0, weight=1)
self.tabview_logs.configure(state='disabled') self.tabview_logs.tab("system").grid_rowconfigure(0, weight=1)
self.tabview_logs.tab("system").grid_columnconfigure(0, weight=1)
# self.tabview_logs.configure(state='disabled')
# add textbox message log # add textbox message send log
self.textbox_message_log = customtkinter.CTkTextbox( self.textbox_message_send_log = customtkinter.CTkTextbox(
self.tabview_logs.tab("send"), self.tabview_logs.tab("send"),
font=customtkinter.CTkFont(family=self.FONT_FAMILY) font=customtkinter.CTkFont(family=self.FONT_FAMILY)
) )
self.textbox_message_log.grid(row=0, column=0, padx=0, pady=0, sticky="nsew") self.textbox_message_send_log.grid(row=0, column=0, padx=0, pady=0, sticky="nsew")
self.textbox_message_log.configure(state='disabled') self.textbox_message_send_log.configure(state='disabled')
# add textbox message log # add textbox message receive log
self.textbox_message_receive_log = customtkinter.CTkTextbox( self.textbox_message_receive_log = customtkinter.CTkTextbox(
self.tabview_logs.tab("receive"), self.tabview_logs.tab("receive"),
font=customtkinter.CTkFont(family=self.FONT_FAMILY) font=customtkinter.CTkFont(family=self.FONT_FAMILY)
) )
self.textbox_message_receive_log.grid(row=0, column=0, padx=0, pady=0, sticky="nsew") self.textbox_message_receive_log.grid(row=0, column=0, padx=0, pady=0, sticky="nsew")
self.textbox_message_receive_log.configure(state='disabled') # self.textbox_message_receive_log.configure(state='disabled')
# add textbox message system log
self.textbox_message_system_log = customtkinter.CTkTextbox(
self.tabview_logs.tab("system"),
font=customtkinter.CTkFont(family=self.FONT_FAMILY)
)
self.textbox_message_system_log.grid(row=0, column=0, padx=0, pady=0, sticky="nsew")
# self.textbox_message_system_log.configure(state='disabled')
# add entry message box # add entry message box
self.entry_message_box = customtkinter.CTkEntry( self.entry_message_box = customtkinter.CTkEntry(
@@ -252,10 +268,10 @@ class App(customtkinter.CTk):
self.translator = translation.Translator() self.translator = translation.Translator()
if self.translator.authentication(self.CHOICE_TRANSLATOR, self.AUTH_KEYS[self.CHOICE_TRANSLATOR]) is False: if self.translator.authentication(self.CHOICE_TRANSLATOR, self.AUTH_KEYS[self.CHOICE_TRANSLATOR]) is False:
# error update Auth key # error update Auth key
self.textbox_message_log.configure(state='normal') self.textbox_message_system_log.configure(state='normal')
self.textbox_message_log.insert("end", f"[ERROR] Auth Keyを設定してないか間違っています\n") self.textbox_message_system_log.insert("end", f"[ERROR] Auth Keyを設定してないか間違っています\n")
self.textbox_message_log.configure(state='disabled') self.textbox_message_system_log.configure(state='disabled')
self.textbox_message_log.see("end") self.textbox_message_system_log.see("end")
## set transcription instance ## set transcription instance
self.vr = transcription.VoiceRecognizer() self.vr = transcription.VoiceRecognizer()
@@ -301,7 +317,10 @@ class App(customtkinter.CTk):
def button_config_callback(self): def button_config_callback(self):
if self.config_window is None or not self.config_window.winfo_exists(): if self.config_window is None or not self.config_window.winfo_exists():
try:
self.config_window = window_config.ToplevelWindowConfig(self) self.config_window = window_config.ToplevelWindowConfig(self)
except Exception as e:
print(e)
self.config_window.focus() self.config_window.focus()
def button_information_callback(self): def button_information_callback(self):
@@ -311,30 +330,32 @@ class App(customtkinter.CTk):
def checkbox_translation_callback(self): def checkbox_translation_callback(self):
self.ENABLE_TRANSLATION = self.checkbox_translation.get() self.ENABLE_TRANSLATION = self.checkbox_translation.get()
self.textbox_message_log.configure(state='normal') self.textbox_message_system_log.configure(state='normal')
if self.ENABLE_TRANSLATION: if self.ENABLE_TRANSLATION:
self.textbox_message_log.insert("end", f"[INFO] start translation\n") self.textbox_message_system_log.insert("end", f"[INFO] start translation\n")
else: else:
self.textbox_message_log.insert("end", f"[INFO] stop translation\n") self.textbox_message_system_log.insert("end", f"[INFO] stop translation\n")
self.textbox_message_log.configure(state='disabled') self.textbox_message_system_log.configure(state='disabled')
self.textbox_message_log.see("end") self.textbox_message_system_log.see("end")
utils.save_json(self.PATH_CONFIG, "ENABLE_TRANSLATION", self.ENABLE_TRANSLATION) utils.save_json(self.PATH_CONFIG, "ENABLE_TRANSLATION", self.ENABLE_TRANSLATION)
def checkbox_transcription_callback(self): def checkbox_transcription_callback(self):
self.ENABLE_TRANSCRIPTION = self.checkbox_transcription.get() self.ENABLE_TRANSCRIPTION = self.checkbox_transcription.get()
if self.ENABLE_TRANSCRIPTION is True: if self.ENABLE_TRANSCRIPTION is True:
# start threading # start threading
self.vr.set_mic(self.CHOICE_MIC_DEVICE) self.vr.set_mic(self.CHOICE_MIC_DEVICE, threshold=self.INPUT_MIC_THRESHOLD, is_dynamic=self.INPUT_MIC_IS_DYNAMIC)
self.vr.init_mic(threshold=self.MIC_THRESHOLD, is_dynamic=self.ENABLE_MIC_IS_DYNAMIC) self.vr.init_mic()
th_vr_listen_mic = threading.Thread(target = self.vr_listen_mic)
th_vr_recognize_mic = threading.Thread(target = self.vr_recognize_mic)
th_vr_listen_mic.start()
th_vr_recognize_mic.start()
self.vr.set_spk(self.CHOICE_SPEAKER_DEVICE) self.vr.set_spk(self.CHOICE_SPEAKER_DEVICE)
self.vr.init_spk() self.vr.init_spk()
th_vr_listen_mic = threading.Thread(target = self.vr_listen_mic)
th_vr_recognize_mic = threading.Thread(target = self.vr_recognize_mic)
th_vr_listen_spk = threading.Thread(target = self.vr_listen_spk) th_vr_listen_spk = threading.Thread(target = self.vr_listen_spk)
th_vr_recognize_spk = threading.Thread(target = self.vr_recognize_spk) th_vr_recognize_spk = threading.Thread(target = self.vr_recognize_spk)
th_vr_listen_mic.start()
th_vr_recognize_mic.start()
th_vr_listen_spk.start() th_vr_listen_spk.start()
th_vr_recognize_spk.start() th_vr_recognize_spk.start()
@@ -345,10 +366,10 @@ class App(customtkinter.CTk):
self.vr.listen_mic() self.vr.listen_mic()
def vr_recognize_mic(self): def vr_recognize_mic(self):
self.textbox_message_log.configure(state='normal') self.textbox_message_system_log.configure(state='normal')
self.textbox_message_log.insert("end", f"[INFO] start transcription\n") self.textbox_message_system_log.insert("end", f"[INFO] start transcription\n")
self.textbox_message_log.configure(state='disabled') self.textbox_message_system_log.configure(state='disabled')
self.textbox_message_log.see("end") self.textbox_message_system_log.see("end")
while self.checkbox_transcription.get() is True: while self.checkbox_transcription.get() is True:
message = self.vr.recognize_mic(language=self.INPUT_MIC_VOICE_LANGUAGE) message = self.vr.recognize_mic(language=self.INPUT_MIC_VOICE_LANGUAGE)
@@ -357,10 +378,10 @@ class App(customtkinter.CTk):
if self.checkbox_translation.get() is False: if self.checkbox_translation.get() is False:
voice_message = f"{message}" voice_message = f"{message}"
elif self.translator.translator_status[self.CHOICE_TRANSLATOR] is False: elif self.translator.translator_status[self.CHOICE_TRANSLATOR] is False:
self.textbox_message_log.configure(state='normal') self.textbox_message_system_log.configure(state='normal')
self.textbox_message_log.insert("end", f"[ERROR] Auth Keyもしくは言語の設定が間違っています\n") self.textbox_message_system_log.insert("end", f"[ERROR] Auth Keyもしくは言語の設定が間違っています\n")
self.textbox_message_log.configure(state='disabled') self.textbox_message_system_log.configure(state='disabled')
self.textbox_message_log.see("end") self.textbox_message_system_log.see("end")
voice_message = f"{message}" voice_message = f"{message}"
else: else:
result = self.translator.translate( result = self.translator.translate(
@@ -375,10 +396,10 @@ class App(customtkinter.CTk):
osc_tools.send_message(voice_message, self.OSC_IP_ADDRESS, self.OSC_PORT) osc_tools.send_message(voice_message, self.OSC_IP_ADDRESS, self.OSC_PORT)
# update textbox message log # update textbox message log
self.textbox_message_log.configure(state='normal') self.textbox_message_send_log.configure(state='normal')
self.textbox_message_log.insert("end", f"[VOICE] {voice_message}\n") self.textbox_message_send_log.insert("end", f"[VOICE] {voice_message}\n")
self.textbox_message_log.configure(state='disabled') self.textbox_message_send_log.configure(state='disabled')
self.textbox_message_log.see("end") self.textbox_message_send_log.see("end")
def vr_listen_spk(self): def vr_listen_spk(self):
while self.checkbox_transcription.get() is True: while self.checkbox_transcription.get() is True:
@@ -392,10 +413,10 @@ class App(customtkinter.CTk):
if self.checkbox_translation.get() is False: if self.checkbox_translation.get() is False:
voice_message = f"{message}" voice_message = f"{message}"
elif self.translator.translator_status[self.CHOICE_TRANSLATOR] is False: elif self.translator.translator_status[self.CHOICE_TRANSLATOR] is False:
self.textbox_message_log.configure(state='normal') self.textbox_message_system_log.configure(state='normal')
self.textbox_message_log.insert("end", f"[ERROR] Auth Keyもしくは言語の設定が間違っています\n") self.textbox_message_system_log.insert("end", f"[ERROR] Auth Keyもしくは言語の設定が間違っています\n")
self.textbox_message_log.configure(state='disabled') self.textbox_message_system_log.configure(state='disabled')
self.textbox_message_log.see("end") self.textbox_message_system_log.see("end")
voice_message = f"{message}" voice_message = f"{message}"
else: else:
result = self.translator.translate( result = self.translator.translate(
@@ -409,11 +430,11 @@ class App(customtkinter.CTk):
# send OSC message # send OSC message
osc_tools.send_message(voice_message, self.OSC_IP_ADDRESS, self.OSC_PORT) osc_tools.send_message(voice_message, self.OSC_IP_ADDRESS, self.OSC_PORT)
# update textbox message log # update textbox message receive log
self.textbox_message_log.configure(state='normal') self.textbox_message_receive_log.configure(state='normal')
self.textbox_message_log.insert("end", f"[VOICE] {voice_message}\n") self.textbox_message_receive_log.insert("end", f"[VOICE] {voice_message}\n")
self.textbox_message_log.configure(state='disabled') self.textbox_message_receive_log.configure(state='disabled')
self.textbox_message_log.see("end") self.textbox_message_receive_log.see("end")
def checkbox_foreground_callback(self): def checkbox_foreground_callback(self):
self.ENABLE_FOREGROUND = self.checkbox_foreground.get() self.ENABLE_FOREGROUND = self.checkbox_foreground.get()
@@ -436,10 +457,10 @@ class App(customtkinter.CTk):
if self.checkbox_translation.get() is False: if self.checkbox_translation.get() is False:
chat_message = f"{message}" chat_message = f"{message}"
elif (self.translator.translator_status[self.CHOICE_TRANSLATOR] is False) or (self.INPUT_SOURCE_LANG == "None") or (self.INPUT_TARGET_LANG == "None"): elif (self.translator.translator_status[self.CHOICE_TRANSLATOR] is False) or (self.INPUT_SOURCE_LANG == "None") or (self.INPUT_TARGET_LANG == "None"):
self.textbox_message_log.configure(state='normal') self.textbox_message_system_log.configure(state='normal')
self.textbox_message_log.insert("end", f"[ERROR] Auth Keyもしくは言語の設定が間違っています\n") self.textbox_message_system_log.insert("end", f"[ERROR] Auth Keyもしくは言語の設定が間違っています\n")
self.textbox_message_log.configure(state='disabled') self.textbox_message_system_log.configure(state='disabled')
self.textbox_message_log.see("end") self.textbox_message_system_log.see("end")
chat_message = f"{message}" chat_message = f"{message}"
else: else:
result = self.translator.translate( result = self.translator.translate(
@@ -451,13 +472,13 @@ class App(customtkinter.CTk):
chat_message = self.MESSAGE_FORMAT.replace("[message]", message).replace("[translation]", result) chat_message = self.MESSAGE_FORMAT.replace("[message]", message).replace("[translation]", result)
# send OSC message # send OSC message
osc_tools.send_message(chat_message, self.OSC_IP_ADDRESS, self.OSC_PORT) # osc_tools.send_message(chat_message, self.OSC_IP_ADDRESS, self.OSC_PORT)
# update textbox message log # update textbox message log
self.textbox_message_log.configure(state='normal') self.textbox_message_receive_log.configure(state='normal')
self.textbox_message_log.insert("end", f"[CHAT] {chat_message}\n") self.textbox_message_receive_log.insert("end", f"[CHAT] {chat_message}\n")
self.textbox_message_log.configure(state='disabled') self.textbox_message_receive_log.configure(state='disabled')
self.textbox_message_log.see("end") self.textbox_message_receive_log.see("end")
# delete message in entry message box # delete message in entry message box
# self.entry_message_box.delete(0, customtkinter.END) # self.entry_message_box.delete(0, customtkinter.END)

54
transcription.py
View File

@@ -20,18 +20,18 @@ class VoiceRecognizer():
"es-NI","es-PA","es-PY","es-PE","es-PR","es-ES","es-UY","es-US","es-VE","sv-SE","th-TH","tr-TR","uk-UA", "es-NI","es-PA","es-PY","es-PE","es-PR","es-ES","es-UY","es-US","es-VE","sv-SE","th-TH","tr-TR","uk-UA",
"vi-VN","zu-ZA" "vi-VN","zu-ZA"
] ]
self.mic = None self.mic_device_name = None
self.enable_mic_recognize = False self.mic_threshold = 50
self.queue_mic = queue.Queue() self.mic_is_dynamic = False
self.mic_queue = queue.Queue()
self.spk_device_name = None self.spk_device_name = None
self.spk_sample_rate = 16000 self.spk_sample_rate = 16000
self.spk_interval = 3 self.spk_interval = 3
self.spk_buffer_size = 4096 self.spk_buffer_size = 4096
self.spk_language = "en-US"
self.spk_audio = np.empty(self.spk_sample_rate * self.spk_interval + self.spk_buffer_size, dtype=np.float32) self.spk_audio = np.empty(self.spk_sample_rate * self.spk_interval + self.spk_buffer_size, dtype=np.float32)
self.n = 0 self.n = 0
self.queue_spk = queue.Queue() self.spk_queue = queue.Queue()
def search_input_device(self): def search_input_device(self):
device_list = sd.query_devices() device_list = sd.query_devices()
@@ -61,48 +61,34 @@ class VoiceRecognizer():
def set_mic(self, device_name, threshold=50, is_dynamic=False): def set_mic(self, device_name, threshold=50, is_dynamic=False):
input_device_list = self.search_input_device() input_device_list = self.search_input_device()
if device_name in [input_device["name"] for input_device in input_device_list]: self.mic_device_name = [device["index"] for device in input_device_list if device["name"] == device_name][0]
index = [device["index"] for device in input_device_list if device["name"] == device_name][0] self.mic_threshold = threshold
self.mic_is_dynamic = is_dynamic
self.mic = sr.Microphone(device_index=index) def init_mic(self):
self.r.energy_threshold = threshold self.r.energy_threshold = self.mic_threshold
if is_dynamic: if self.mic_is_dynamic:
with self.mic as source: with self.mic as source:
self.r.adjust_for_ambient_noise(source, 3.0) self.r.adjust_for_ambient_noise(source, 3.0)
return True
else:
return False
def init_mic(self, threshold=50, is_dynamic=False):
if isinstance(self.mic, sr.Microphone):
self.r.energy_threshold = threshold
if is_dynamic:
with self.mic as source:
self.r.adjust_for_ambient_noise(source, 3.0)
return True
else:
return False
def listen_mic(self): def listen_mic(self):
if self.mic != None: with sr.Microphone(device_index=self.mic_device_name) as source:
with self.mic as source:
audio = self.r.listen(source) audio = self.r.listen(source)
self.queue_mic.put(audio) self.mic_queue.put(audio)
def recognize_mic(self, language): def recognize_mic(self, language):
try: try:
audio = self.queue_mic.get() audio = self.mic_queue.get()
text = self.r.recognize_google(audio, language=language) text = self.r.recognize_google(audio, language=language)
except: except:
text = "" text = ""
return text return text
def set_spk(self, device_name=str(sc.default_speaker().name), sample_rate=16000, interval=3, buffer_size=4096, language="en-US"): def set_spk(self, device_name=str(sc.default_speaker().name), sample_rate=16000, interval=3, buffer_size=4096):
self.spk_device_name = device_name self.spk_device_name = device_name
self.spk_sample_rate = sample_rate self.spk_sample_rate = sample_rate
self.spk_interval = interval self.spk_interval = interval
self.spk_buffer_size = buffer_size self.spk_buffer_size = buffer_size
self.spk_language = language
def init_spk(self): def init_spk(self):
self.spk_audio = np.empty(self.spk_sample_rate * self.spk_interval + self.spk_buffer_size, dtype=np.float32) self.spk_audio = np.empty(self.spk_sample_rate * self.spk_interval + self.spk_buffer_size, dtype=np.float32)
@@ -120,23 +106,23 @@ class VoiceRecognizer():
vol = np.convolve(audio[m:n] ** 2, np.ones(100) / 100, 'same') vol = np.convolve(audio[m:n] ** 2, np.ones(100) / 100, 'same')
m += vol.argmin() m += vol.argmin()
audio_prev = audio.copy() audio_prev = audio.copy()
self.queue_spk.put(audio[:m]) self.spk_queue.put(audio[:m])
audio = np.empty(self.spk_sample_rate * self.spk_interval + self.spk_buffer_size, dtype=np.float32) audio = np.empty(self.spk_sample_rate * self.spk_interval + self.spk_buffer_size, dtype=np.float32)
audio[:n-m] = audio_prev[m:n] audio[:n-m] = audio_prev[m:n]
n = n-m n = n-m
self.spk_audio = audio self.spk_audio = audio
self.n = n self.n = n
def recognize_spk(self): def recognize_spk(self, language):
try: try:
audio = self.queue_spk.get() audio = self.spk_queue.get()
with io.BytesIO() as memory_file: with io.BytesIO() as memory_file:
sf.write(file=memory_file, data=audio, format="WAV", samplerate=self.spk_sample_rate) sf.write(file=memory_file, data=audio, format="WAV", samplerate=self.spk_sample_rate)
memory_file.seek(0) memory_file.seek(0)
with sr.AudioFile(memory_file) as source: with sr.AudioFile(memory_file) as source:
audio = self.r.record(source) audio = self.r.record(source)
text = self.r.recognize_google(audio, language=self.spk_language) text = self.r.recognize_google(audio, language=language)
except Exception as e: except:
text = "" text = ""
return text return text

132
window_config.py
View File

@@ -245,7 +245,7 @@ class ToplevelWindowConfig(customtkinter.CTkToplevel):
font=customtkinter.CTkFont(family=self.parent.FONT_FAMILY) font=customtkinter.CTkFont(family=self.parent.FONT_FAMILY)
) )
self.checkbox_input_mic_is_dynamic.grid(row=2, column=1, columnspan=3 ,padx=5, pady=5, sticky="nsew") self.checkbox_input_mic_is_dynamic.grid(row=2, column=1, columnspan=3 ,padx=5, pady=5, sticky="nsew")
if self.parent.ENABLE_MIC_IS_DYNAMIC is True: if self.parent.INPUT_MIC_IS_DYNAMIC is True:
self.checkbox_input_mic_is_dynamic.select() self.checkbox_input_mic_is_dynamic.select()
else: else:
self.checkbox_input_mic_is_dynamic.deselect() self.checkbox_input_mic_is_dynamic.deselect()
@@ -263,7 +263,7 @@ class ToplevelWindowConfig(customtkinter.CTkToplevel):
from_=0, from_=0,
to=300, to=300,
command=self.slider_input_mic_threshold_callback, command=self.slider_input_mic_threshold_callback,
variable=tk.DoubleVar(value=self.parent.MIC_THRESHOLD), variable=tk.DoubleVar(value=self.parent.INPUT_MIC_THRESHOLD),
) )
self.slider_input_mic_threshold.grid(row=3, column=1, columnspan=3 ,padx=5, pady=10, sticky="nsew") self.slider_input_mic_threshold.grid(row=3, column=1, columnspan=3 ,padx=5, pady=10, sticky="nsew")
@@ -272,16 +272,15 @@ class ToplevelWindowConfig(customtkinter.CTkToplevel):
self.tabview_config.tab("Transcription"), self.tabview_config.tab("Transcription"),
text="Input Speaker Device:", text="Input Speaker Device:",
fg_color="transparent", fg_color="transparent",
font=customtkinter.CTkFont(family=self.parent.FONT_FAMILY, overstrike=True) font=customtkinter.CTkFont(family=self.parent.FONT_FAMILY)
) )
self.label_input_speaker_device.grid(row=4, column=0, columnspan=1, padx=5, pady=5, sticky="nsw") self.label_input_speaker_device.grid(row=4, column=0, columnspan=1, padx=5, pady=5, sticky="nsw")
self.optionmenu_input_speaker_device = customtkinter.CTkOptionMenu( self.optionmenu_input_speaker_device = customtkinter.CTkOptionMenu(
self.tabview_config.tab("Transcription"), self.tabview_config.tab("Transcription"),
values=self.parent.vr.search_output_device(), values=self.parent.vr.search_output_device(),
# command=self.optionmenu_input_speaker_device_callback, command=self.optionmenu_input_speaker_device_callback,
font=customtkinter.CTkFont(family=self.parent.FONT_FAMILY), font=customtkinter.CTkFont(family=self.parent.FONT_FAMILY),
variable=customtkinter.StringVar(value=self.parent.CHOICE_SPEAKER_DEVICE), variable=customtkinter.StringVar(value=self.parent.CHOICE_SPEAKER_DEVICE),
state="disabled"
) )
self.optionmenu_input_speaker_device.grid(row=4, column=1, columnspan=3 ,padx=5, pady=5, sticky="nsew") self.optionmenu_input_speaker_device.grid(row=4, column=1, columnspan=3 ,padx=5, pady=5, sticky="nsew")
@@ -290,59 +289,65 @@ class ToplevelWindowConfig(customtkinter.CTkToplevel):
self.tabview_config.tab("Transcription"), self.tabview_config.tab("Transcription"),
text="Input Speaker Voice Language:", text="Input Speaker Voice Language:",
fg_color="transparent", fg_color="transparent",
font=customtkinter.CTkFont(family=self.parent.FONT_FAMILY, overstrike=True) font=customtkinter.CTkFont(family=self.parent.FONT_FAMILY)
) )
self.label_input_speaker_voice_language.grid(row=5, column=0, columnspan=1, padx=5, pady=5, sticky="nsw") self.label_input_speaker_voice_language.grid(row=5, column=0, columnspan=1, padx=5, pady=5, sticky="nsw")
self.optionmenu_input_speaker_voice_language = customtkinter.CTkOptionMenu( self.optionmenu_input_speaker_voice_language = customtkinter.CTkOptionMenu(
self.tabview_config.tab("Transcription"), self.tabview_config.tab("Transcription"),
values=list(self.parent.vr.languages), values=list(self.parent.vr.languages),
# command=self.optionmenu_input_speaker_voice_language_callback, command=self.optionmenu_input_speaker_voice_language_callback,
font=customtkinter.CTkFont(family=self.parent.FONT_FAMILY), font=customtkinter.CTkFont(family=self.parent.FONT_FAMILY),
variable=customtkinter.StringVar(value=self.parent.INPUT_SPEAKER_VOICE_LANGUAGE), variable=customtkinter.StringVar(value=self.parent.INPUT_SPEAKER_VOICE_LANGUAGE),
state="disabled"
) )
self.optionmenu_input_speaker_voice_language.grid(row=5, column=1, columnspan=3 ,padx=5, pady=5, sticky="nsew") self.optionmenu_input_speaker_voice_language.grid(row=5, column=1, columnspan=3 ,padx=5, pady=5, sticky="nsew")
## checkbox input speaker in dynamic ## entry input speaker sampling rate
self.label_input_speaker_is_dynamic = customtkinter.CTkLabel( self.label_input_speaker_sampling_rate = customtkinter.CTkLabel(
self.tabview_config.tab("Transcription"), self.tabview_config.tab("Transcription"),
text="Input Speaker IsDynamic:", text="Input Speaker SamplingRate:",
fg_color="transparent", fg_color="transparent",
font=customtkinter.CTkFont(family=self.parent.FONT_FAMILY, overstrike=True) font=customtkinter.CTkFont(family=self.parent.FONT_FAMILY)
) )
self.label_input_speaker_is_dynamic.grid(row=6, column=0, columnspan=1, padx=5, pady=5, sticky="nsw") self.label_input_speaker_sampling_rate.grid(row=6, column=0, columnspan=1, padx=5, pady=5, sticky="nsw")
self.checkbox_input_speaker_is_dynamic = customtkinter.CTkCheckBox( self.entry_input_speaker_sampling_rate = customtkinter.CTkEntry(
self.tabview_config.tab("Transcription"), self.tabview_config.tab("Transcription"),
text="", textvariable=customtkinter.StringVar(value=self.parent.INPUT_SPEAKER_SAMPLING_RATE),
onvalue=True, font=customtkinter.CTkFont(family=self.parent.FONT_FAMILY)
offvalue=False,
# command=self.checkbox_input_speaker_is_dynamic_callback,
font=customtkinter.CTkFont(family=self.parent.FONT_FAMILY),
state="disabled"
) )
self.checkbox_input_speaker_is_dynamic.grid(row=6, column=1, columnspan=3 ,padx=5, pady=5, sticky="nsew") self.entry_input_speaker_sampling_rate.grid(row=6, column=1, columnspan=3 ,padx=5, pady=5, sticky="nsew")
if self.parent.ENABLE_SPEAKER_IS_DYNAMIC is True: self.entry_input_speaker_sampling_rate.bind("<Any-KeyRelease>", self.entry_input_speaker_sampling_rate_callback)
self.checkbox_input_speaker_is_dynamic.select()
else:
self.checkbox_input_speaker_is_dynamic.deselect()
## slider input speaker threshold ## entry input speaker interval
self.label_input_speaker_threshold = customtkinter.CTkLabel( self.label_input_speaker_interval = customtkinter.CTkLabel(
self.tabview_config.tab("Transcription"), self.tabview_config.tab("Transcription"),
text="Input Speaker Threshold:", text="Input Speaker Interval:",
fg_color="transparent", fg_color="transparent",
font=customtkinter.CTkFont(family=self.parent.FONT_FAMILY, overstrike=True) font=customtkinter.CTkFont(family=self.parent.FONT_FAMILY)
) )
self.label_input_speaker_threshold.grid(row=7, column=0, columnspan=1, padx=5, pady=5, sticky="nsw") self.label_input_speaker_interval.grid(row=7, column=0, columnspan=1, padx=5, pady=5, sticky="nsw")
self.slider_input_speaker_threshold = customtkinter.CTkSlider( self.entry_input_speaker_interval = customtkinter.CTkEntry(
self.tabview_config.tab("Transcription"), self.tabview_config.tab("Transcription"),
from_=0, textvariable=customtkinter.StringVar(value=self.parent.INPUT_SPEAKER_INTERVAL),
to=300, font=customtkinter.CTkFont(family=self.parent.FONT_FAMILY)
# command=self.slider_input_speaker_threshold_callback,
variable=tk.DoubleVar(value=self.parent.SPEAKER_THRESHOLD),
state="disabled"
) )
self.slider_input_speaker_threshold.grid(row=7, column=1, columnspan=3 ,padx=5, pady=10, sticky="nsew") self.entry_input_speaker_interval.grid(row=7, column=1, columnspan=3 ,padx=5, pady=5, sticky="nsew")
self.entry_input_speaker_interval.bind("<Any-KeyRelease>", self.entry_input_speaker_interval_callback)
## entry input speaker buffer size
self.label_input_speaker_buffer_size = customtkinter.CTkLabel(
self.tabview_config.tab("Transcription"),
text="Input Speaker BufferSize:",
fg_color="transparent",
font=customtkinter.CTkFont(family=self.parent.FONT_FAMILY)
)
self.label_input_speaker_buffer_size.grid(row=8, column=0, columnspan=1, padx=5, pady=5, sticky="nsw")
self.entry_input_speaker_buffer_size = customtkinter.CTkEntry(
self.tabview_config.tab("Transcription"),
textvariable=customtkinter.StringVar(value=self.parent.INPUT_SPEAKER_BUFFER_SIZE),
font=customtkinter.CTkFont(family=self.parent.FONT_FAMILY)
)
self.entry_input_speaker_buffer_size.grid(row=8, column=1, columnspan=3 ,padx=5, pady=5, sticky="nsew")
self.entry_input_speaker_buffer_size.bind("<Any-KeyRelease>", self.entry_input_speaker_buffer_size_callback)
# tab Parameter # tab Parameter
## entry ip address ## entry ip address
@@ -522,10 +527,10 @@ class ToplevelWindowConfig(customtkinter.CTkToplevel):
def optionmenu_translation_translator_callback(self, choice): def optionmenu_translation_translator_callback(self, choice):
if self.parent.translator.authentication(choice, self.parent.AUTH_KEYS[choice]) is False: if self.parent.translator.authentication(choice, self.parent.AUTH_KEYS[choice]) is False:
self.parent.textbox_message_log.configure(state='normal') self.parent.textbox_message_system_log.configure(state='normal')
self.parent.textbox_message_log.insert("end", f"[ERROR]Auth Keyを設定してないか間違っています\n") self.parent.textbox_message_system_log.insert("end", f"[ERROR]Auth Keyを設定してないか間違っています\n")
self.parent.textbox_message_log.configure(state='disabled') self.parent.textbox_message_system_log.configure(state='disabled')
self.parent.textbox_message_log.see("end") self.parent.textbox_message_system_log.see("end")
else: else:
self.optionmenu_translation_input_source_language.configure( self.optionmenu_translation_input_source_language.configure(
values=self.parent.translator.languages[choice], values=self.parent.translator.languages[choice],
@@ -559,14 +564,34 @@ class ToplevelWindowConfig(customtkinter.CTkToplevel):
def checkbox_input_mic_is_dynamic_callback(self): def checkbox_input_mic_is_dynamic_callback(self):
value = self.checkbox_input_mic_is_dynamic.get() value = self.checkbox_input_mic_is_dynamic.get()
self.parent.ENABLE_MIC_IS_DYNAMIC = value self.parent.INPUT_MIC_IS_DYNAMIC = value
utils.save_json(self.parent.PATH_CONFIG, "ENABLE_MIC_IS_DYNAMIC", self.parent.ENABLE_MIC_IS_DYNAMIC) utils.save_json(self.parent.PATH_CONFIG, "INPUT_MIC_IS_DYNAMIC", self.parent.INPUT_MIC_IS_DYNAMIC)
self.parent.vr.init_mic(threshold=self.parent.MIC_THRESHOLD, is_dynamic=self.parent.ENABLE_MIC_IS_DYNAMIC) self.parent.vr.set_mic(self.parent.CHOICE_MIC_DEVICE, threshold=self.parent.INPUT_MIC_THRESHOLD, is_dynamic=self.parent.INPUT_MIC_IS_DYNAMIC)
def slider_input_mic_threshold_callback(self, value): def slider_input_mic_threshold_callback(self, value):
self.parent.MIC_THRESHOLD = value self.parent.INPUT_MIC_THRESHOLD = value
utils.save_json(self.parent.PATH_CONFIG, "MIC_THRESHOLD", self.parent.MIC_THRESHOLD) utils.save_json(self.parent.PATH_CONFIG, "INPUT_MIC_THRESHOLD", self.parent.INPUT_MIC_THRESHOLD)
self.parent.vr.init_mic(threshold=self.parent.MIC_THRESHOLD, is_dynamic=self.parent.ENABLE_MIC_IS_DYNAMIC) self.parent.vr.set_mic(self.parent.CHOICE_MIC_DEVICE, threshold=self.parent.INPUT_MIC_THRESHOLD, is_dynamic=self.parent.INPUT_MIC_IS_DYNAMIC)
def optionmenu_input_speaker_device_callback(self, choice):
self.parent.CHOICE_SPEAKER_DEVICE = choice
utils.save_json(self.parent.PATH_CONFIG, "CHOICE_SPEAKER_DEVICE", self.parent.CHOICE_SPEAKER_DEVICE)
def optionmenu_input_speaker_voice_language_callback(self, choice):
self.parent.INPUT_SPEAKER_VOICE_LANGUAGE = choice
utils.save_json(self.parent.PATH_CONFIG, "INPUT_SPEAKER_VOICE_LANGUAGE", self.parent.INPUT_SPEAKER_VOICE_LANGUAGE)
def entry_input_speaker_sampling_rate_callback(self, event):
self.parent.INPUT_SPEAKER_SAMPLING_RATE = int(self.entry_input_speaker_sampling_rate.get())
utils.save_json(self.parent.PATH_CONFIG, "INPUT_SPEAKER_SAMPLING_RATE", self.parent.INPUT_SPEAKER_SAMPLING_RATE)
def entry_input_speaker_interval_callback(self, event):
self.parent.INPUT_SPEAKER_INTERVAL = int(self.entry_input_speaker_interval.get())
utils.save_json(self.parent.PATH_CONFIG, "INPUT_SPEAKER_INTERVAL", self.parent.INPUT_SPEAKER_INTERVAL)
def entry_input_speaker_buffer_size_callback(self, event):
self.parent.INPUT_SPEAKER_BUFFER_SIZE = int(self.entry_input_speaker_buffer_size.get())
utils.save_json(self.parent.PATH_CONFIG, "INPUT_SPEAKER_BUFFER_SIZE", self.parent.INPUT_SPEAKER_BUFFER_SIZE)
def update_ip_address(self): def update_ip_address(self):
value = self.entry_ip_address.get() value = self.entry_ip_address.get()
@@ -583,19 +608,14 @@ class ToplevelWindowConfig(customtkinter.CTkToplevel):
def update_authkey(self): def update_authkey(self):
value = self.entry_authkey.get() value = self.entry_authkey.get()
if len(value) > 0: if len(value) > 0:
self.parent.textbox_message_log.configure(state='normal')
self.parent.textbox_message_log.delete("0.0", "end")
self.parent.textbox_message_log.configure(state='disabled')
self.parent.textbox_message_log.see("end")
if self.parent.translator.authentication(self.parent.CHOICE_TRANSLATOR, self.parent.AUTH_KEYS[self.parent.CHOICE_TRANSLATOR]) is True: if self.parent.translator.authentication(self.parent.CHOICE_TRANSLATOR, self.parent.AUTH_KEYS[self.parent.CHOICE_TRANSLATOR]) is True:
self.parent.AUTH_KEYS["DeepL(auth)"] = value self.parent.AUTH_KEYS["DeepL(auth)"] = value
utils.save_json(self.parent.PATH_CONFIG, "AUTH_KEYS", self.parent.AUTH_KEYS) utils.save_json(self.parent.PATH_CONFIG, "AUTH_KEYS", self.parent.AUTH_KEYS)
else: else:
self.parent.textbox_message_log.configure(state='normal') self.parent.textbox_message_system_log.configure(state='normal')
self.parent.textbox_message_log.insert("end", f"[ERROR]Auth Keyを設定してないか間違っています\n") self.parent.textbox_message_system_log.insert("end", f"[ERROR]Auth Keyを設定してないか間違っています\n")
self.parent.textbox_message_log.configure(state='disabled') self.parent.textbox_message_system_log.configure(state='disabled')
self.parent.textbox_message_log.see("end") self.parent.textbox_message_system_log.see("end")
def update_message_format(self): def update_message_format(self):
value = self.entry_message_format.get() value = self.entry_message_format.get()