def __init__(self) -> None:
"""
Assign the speech recognition object and start the audio handling thread
"""
self.__listener = sr.Recognizer()
__audio_proc = Thread(target=self.__outputHandler)
__audio_proc.start()
# Remove stragglers to avoid corruption exceptions
temp_storage_dir = path.dirname(Configuration.GetFilePath("temp_microphone_storage"))
for file in listdir(temp_storage_dir):
remove(os.path.join(path.dirname(Configuration.GetFilePath("temp_microphone_storage")), file))
def setOutput(self, text: str) -> None:
"""
Set audio output for handler to speak
:param text: output text
:return: None
"""
self.__output_stack.append(text)
base_dir = path.dirname(Configuration.GetFilePath("temp_microphone_storage"))
output_path = Configuration.GetFilePath("temp_microphone_storage").format(audio_id=len([file for file in
listdir(base_dir) if
path.isfile(
path.join(base_dir,
file))]))
language = Configuration.Get("language")
out = gTTS(text, lang=language)
out.save(output_path)
def start(self, render: bool = True) -> None:
"""
Currently doesnt do a lot except for starting render,
placeholder if more startup functionality is required.
"""
if render:
self.__render(Configuration.GetFilePath("video_output_location"))
def __outputHandler(self) -> None:
"""
Output handling thread
:return: None
"""
base_dir = path.dirname(Configuration.GetFilePath("temp_microphone_storage"))
while True:
for file in [path.join(base_dir, file) for file in listdir(base_dir) if
path.isfile(path.join(base_dir, file))]:
try:
playsound.playsound(file)
except playsound.PlaysoundException as e:
print(f"Playback error: {e}")
if path.exists(file):
try:
remove(file)
except PermissionError as e:
continue
time.sleep(1)
def __faceTrack(self, frame, encodings, write_process,
output_location) -> cv2.VideoWriter:
"""
Track faces upon a frame
:param frame: target frame
:param encodings: known encodings
:param write_process: (un)active write process
:param output_location: output for writing process
:return: updated writing process
"""
rgb_frame = cv2.cvtColor(frame, cv2.COLOR_BGR2RGB)
locations = face_recognition.face_locations(rgb_frame)
found_names = []
try:
for encoding in face_recognition.face_encodings(
rgb_frame, locations):
name = None
match = None
for set_name in encodings.keys():
match = face_recognition.compare_faces(encodings[set_name],
encoding,
tolerance=0.50)
if match[0]:
name = set_name
break
self.__lastDetectedTime = datetime.now()
found_names.append(name)
except Exception as e:
print("Error parsing face name: ", e)
try:
for (top, right, bottom,
left), name in zip(locations, found_names):
cv2.rectangle(frame, (left, top), (right, bottom), (0, 0, 0),
2)
cv2.rectangle(frame, (left, bottom), (right, bottom),
(0, 0, 0), cv2.FILLED)
font = cv2.FONT_HERSHEY_DUPLEX
name = name if name is not None else "Unkown"
cv2.putText(frame, name, (left + 6, bottom - 6), font, 0.5,
(255, 255, 255), 1)
if self.__extractKnownFaces:
frame_slice = rgb_frame[top:bottom, left:right]
file_path = os.path.join(
Configuration.GetFilePath("know_face_encodings"), name)
if not os.path.exists(file_path):
os.mkdir(file_path)
Image.fromarray(frame_slice).save(
os.path.join(file_path,
f"{len(os.listdir(file_path))}.png"))
except Exception as e:
print("Error setting face rect: ", e)
if locations and write_process is None:
try:
day_dir, timestamp = f'{re.sub(r"[-.:]", "_", str(datetime.now()))}'.split(
" ")
base_dir = os.path.dirname(
os.path.dirname(os.path.abspath(output_location)))
if day_dir not in os.listdir(base_dir):
os.mkdir(
os.path.dirname(
output_location.format(directory=day_dir,
timestamp="")))
write_process = cv2.VideoWriter(
output_location.format(directory=day_dir,
timestamp=timestamp),
cv2.VideoWriter_fourcc(*'XVID'), 30.0, (640, 480))
except Exception as e:
print(e)
elif len(locations) == 0 and write_process is not None:
write_process = None
return write_process