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

update search_input_device search_default_device

Browse Source
This commit is contained in:
misyaguziya
2023-06-26 16:18:35 +09:00
parent aa18458ed5
commit e6e4e3e305
1 changed files with 64 additions and 48 deletions
Download Patch File Download Diff File Expand all files Collapse all files

112
transcription.py
View File

@@ -1,5 +1,4 @@
import queue import queue
import sounddevice as sd
import speech_recognition as sr import speech_recognition as sr
import pyaudiowpatch as pyaudio import pyaudiowpatch as pyaudio
@@ -33,10 +32,13 @@ class VoiceRecognizer():
def search_input_device(self): def search_input_device(self):
devices = [] devices = []
device_list = sd.query_devices() with pyaudio.PyAudio() as p:
for device in device_list: wasapi_info = p.get_host_api_info_by_type(pyaudio.paWASAPI)
if device["max_input_channels"] > 0: for host_index in range(0, p.get_host_api_count()):
devices.append(device) 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)
return devices return devices
def search_output_device(self): def search_output_device(self):
@@ -51,18 +53,28 @@ class VoiceRecognizer():
return devices return devices
def search_default_device(self): def search_default_device(self):
device_list = sd.query_devices()
mic_index = sd.default.device[0]
name_mic = device_list[mic_index]["name"]
with pyaudio.PyAudio() as p: with pyaudio.PyAudio() as p:
wasapi_info = p.get_host_api_info_by_type(pyaudio.paWASAPI) wasapi_info = p.get_host_api_info_by_type(pyaudio.paWASAPI)
default_speakers = p.get_device_info_by_index(wasapi_info["defaultOutputDevice"]) defaultInputDevice, defaultOutputDevice = wasapi_info["defaultInputDevice"], wasapi_info["defaultOutputDevice"]
if not default_speakers["isLoopbackDevice"]: for host_index in range(0, p.get_host_api_count()):
for loopback in p.get_loopback_device_info_generator(): for device_index in range(0, p. get_host_api_info_by_index(host_index)['deviceCount']):
if default_speakers["name"] in loopback["name"]: device = p.get_device_info_by_host_api_device_index(host_index, device_index)
name_spk = loopback["name"] if device["index"] == defaultInputDevice:
default_mics = device
name_mic = default_mics["name"]
break break
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']):
device = p.get_device_info_by_host_api_device_index(host_index, device_index)
if device["index"] == defaultOutputDevice:
default_speakers = device
if not default_speakers["isLoopbackDevice"]:
for loopback in p.get_loopback_device_info_generator():
if default_speakers["name"] in loopback["name"]:
name_spk = loopback["name"]
break
return name_mic, name_spk return name_mic, name_spk
def set_mic(self, device_name, threshold=50, is_dynamic=False, language="ja-JP"): def set_mic(self, device_name, threshold=50, is_dynamic=False, language="ja-JP"):
@@ -141,45 +153,49 @@ class VoiceRecognizer():
return text return text
if __name__ == "__main__": if __name__ == "__main__":
import queue # import queue
import threading # import threading
mic_queue = queue.Queue() # mic_queue = queue.Queue()
spk_queue = queue.Queue() # spk_queue = queue.Queue()
vr = VoiceRecognizer(mic_queue, spk_queue) # vr = VoiceRecognizer(mic_queue, spk_queue)
mic_name, spk_name = vr.search_default_device() # mic_name, spk_name = vr.search_default_device()
print("mic_name", mic_name) # print("mic_name", mic_name)
print("spk_name", spk_name) # print("spk_name", spk_name)
############################################################### # ###############################################################
vr.set_mic(device_name=mic_name, threshold=300, is_dynamic=False, language="ja-JP") # vr.set_mic(device_name=mic_name, threshold=300, is_dynamic=False, language="ja-JP")
vr.init_mic() # vr.init_mic()
def vr_listen_mic(): # def vr_listen_mic():
while True: # while True:
vr.listen_mic() # vr.listen_mic()
def vr_recognize_mic(): # def vr_recognize_mic():
while True: # while True:
text = vr.recognize_mic() # text = vr.recognize_mic()
if len(text) > 0: # if len(text) > 0:
print(text) # print(text)
th_vr_listen_mic = threading.Thread(target=vr_listen_mic) # th_vr_listen_mic = threading.Thread(target=vr_listen_mic)
th_vr_listen_mic.start() # th_vr_listen_mic.start()
th_vr_recognize_mic = threading.Thread(target=vr_recognize_mic) # th_vr_recognize_mic = threading.Thread(target=vr_recognize_mic)
th_vr_recognize_mic.start() # th_vr_recognize_mic.start()
############################################################### # ###############################################################
############################################################### # ###############################################################
vr.set_spk(device_name=spk_name, interval=4, language="ja-JP") # vr.set_spk(device_name=spk_name, interval=4, language="ja-JP")
vr.start_spk_recording() # vr.start_spk_recording()
def vr_recognize_spk(): # def vr_recognize_spk():
while True: # while True:
text = vr.recognize_spk() # text = vr.recognize_spk()
if len(text) > 0: # if len(text) > 0:
print(text) # print(text)
th_vr_recognize_spk = threading.Thread(target=vr_recognize_spk) # th_vr_recognize_spk = threading.Thread(target=vr_recognize_spk)
th_vr_recognize_spk.start() # th_vr_recognize_spk.start()
############################################################### # ###############################################################
vr = VoiceRecognizer()
print(vr.search_input_device())
print(vr.search_default_device())