def gat_with_breakdown(path_to_file, recognizer: sr.Recognizer, breakdown_duration=10, tempfile_name='temp.wav'):
"""
gat = get audio transcript
This breakdown is effective for writing down SRT(s).
"""
for start, end, total in chop_chop(path_to_file, breakdown_duration, output_file=tempfile_name):
last_yield = start
with sr.AudioFile(tempfile_name) as source:
try:
if (content := recognizer.recognize_google(recognizer.record(source))):
yield {'start': format_to_ffmpeg_duration(start), 'end': format_to_ffmpeg_duration(end), 'text': content, 'total': total, 'delta': end - last_yield}
last_yield = end
except sr.UnknownValueError:
pass
def GetTextInAudio(audio_msg, chat_id, bot):
from speech_recognition import AudioFile, Recognizer, UnknownValueError
rec = Recognizer()
file_id = audio_msg['file_id']
tempAudio = dirLoc + file_id + '.' + audio_msg['mime_type'].split('/')[1]
bot.download_file(file_id, tempAudio)
from pydub import AudioSegment
filename = file_id + '.wav'
file_loc = dirLoc + filename
sound = AudioSegment.from_file(tempAudio)
sound.export(file_loc, format='wav')
with AudioFile(file_loc) as AudioSrc:
content = rec.record(AudioSrc)
try:
text = rec.recognize_google(
audio_data=content,
language='ja-JP',
)
except UnknownValueError:
bot.sendMessage(chat_id, 'None')
return
mainText = t_j2k(text)
bot.sendMessage(chat_id, mainText)
remove(tempAudio)
remove(file_loc)
class MicrophoneMonitor(object):
def __init__(self) -> None:
self.micro = Recognizer()
self.micro.energy_threshold = 4000
self.micro.pause_threshold = 0.5
def monitor_microphone(self, hotword='jarvis'):
with Microphone() as source:
self.micro.adjust_for_ambient_noise(source, duration=0.5)
while True:
print('Aguardando comando: ')
audio = self.micro.listen(source)
try:
trigger = self.micro.recognize_google(audio, language='pt')
# with open("microphone-results.wav", "wb") as f:
# f.write(audio.get_wav_data())
print(trigger)
if hotword.lower() in trigger or hotword.capitalize(
) in trigger:
print('Comando: ', trigger)
return trigger
except UnknownValueError:
print(
"Google Speech Recognition could not understand audio")
except RequestError as e:
print(
"Could not request results from Google Speech Recognition service; {0}"
.format(e))
class Ear: # pylint: disable=too-few-public-methods
"""
listen() -- take audio from microphone, recognize it and return as string
"""
def __init__(self):
self.recognizer = Recognizer()
def listen(self):
"""
listen audio from microphone and convert to the string.
:return:data
"""
self.recognizer = Recognizer()
with Microphone() as source:
self.recognizer.adjust_for_ambient_noise(source)
print("I am listening you...")
audio = self.recognizer.listen(source, phrase_time_limit=6)
data = ""
try:
# Uses the default API key
# To use another API key:
# `r.recognize_google(audio, key="GOOGLE_SPEECH_RECOGNITION_API_KEY")`
data = self.recognizer.recognize_google(audio, language="tr")
print("You said : " + data)
except UnknownValueError:
print("Google Speech Recognition could not understand audio")
return self.listen()
except RequestError as exception:
print("Could not request results from "
"Google Speech Recognition service; {0}".format(exception))
return self.listen()
return data.lower()
def listen(self) -> str:
# Record Audio
data = ''
recognizer = Recognizer()
with Microphone() as source:
self.log.debug('Listening on microphone...')
try:
audio = recognizer.listen(source, timeout=10)
except WaitTimeoutError:
self.log.debug('Listen timed out.')
return data
# Speech recognition using Google Speech Recognition
try:
# Uses the default API key
# To use another API key: `r.recognize_google(audio, key='GOOGLE_SPEECH_RECOGNITION_API_KEY')`
data = recognizer.recognize_google(audio)
self.log.debug(f'You said: "{data}"')
except UnknownValueError:
self.log.debug(
'Google Speech Recognition could not understand audio')
except RequestError as e:
self.log.error(
f'Could not request results from Google Speech Recognition service: {e}'
)
return str(data)
def Audio_file_Read(filename):
universal_dict = {}
cnt = {}
gantu = [0, 0, 0, 0]
analysis = {}
token = Tokenizer()
recog = Recognizer()
try:
audioFile = sr.AudioFile(filename)
with audioFile as source:
audio = recog.record(source)
recognized = recog.recognize_google(audio, language="ko-KR")
res = text_to_word_sequence(recognized)
cnt = collections.Counter(res)
universal_dict = dict(cnt)
if "어" in universal_dict:
gantu[0] = universal_dict["어"]
if "아니" in universal_dict:
gantu[1] = universal_dict["아니"]
if "근데" in universal_dict:
gantu[2] = universal_dict["근데"]
if "이제" in universal_dict:
gantu[3] = universal_dict["이제"]
text = recognized
analysis['text'] = text
analysis['data'] = gantu
return analysis
except UnknownValueError:
analysis['text'] = "당신이 말한 문장이 없습니다."
analysis['data'] = [0, 0, 0, 0]
return analysis
def wav_to_text(wav_file_path, language="es-ES", show_all=False):
r = Recognizer()
with WavFile(wav_file_path) as source:
audio = r.record(source)
try:
return r.recognize_google(audio_data=audio, language=language, show_all=show_all)
except UnknownValueError:
raise GolemException("Could not understand audio")
class SimpleSTT(object):
def __init__(self):
self.recognizer = Recognizer()
def transcribe(self, path_to_source):
with AudioFile(path_to_source) as source:
audio = self.recognizer.listen(source)
return self.recognizer.recognize_google(audio)
def VoiceInput():
recog = Recognizer()
mic = Microphone(device_index=1)
with mic:
audio = recog.listen(mic)
try:
recognized = recog.recognize_google(audio)
except UnknownValueError:
Response("Try Again")
with mic:
audio = recog.listen(mic)
recognized = recog.recognize_google(audio)
except RequestError as exc:
Response("Sorry my service is down")
print(recognized)
return (recognized)
def dothis(message):
"""
From speech to text
:param message:
:return: text
"""
session = message.get_session()
ans = ''
current_cmd = message.get_setting(session, 'active')
if message.attachments['sound']:
try:
r = Recognizer()
mode = 'google'
lang = 'ru-RUS'
ans = ''
for attachment in message.attachments['sound']:
ext = attachment[1]
path = os.path.abspath(os.curdir)
fname = time.strftime("%Y%m%d-%H%M%S") + '.'
dir = path + '/temp/' + fname
urllib.request.urlretrieve(
attachment[0], dir + ext) # getting file
if ext != 'wav':
subprocess.run(['ffmpeg', '-i', dir + ext, dir + 'wav'])
os.remove(dir + ext)
with AudioFile(dir + 'wav') as source:
song = r.record(source)
os.remove(dir + 'wav')
if "en" in message.params:
lang = 'en-EN'
if 'wit' in message.params:
mode = 'wit'
recg = r.recognize_google(
song,
language=lang
) if mode == 'google' else r.recognize_wit(song, witkey)
ans += f">>>>>>{recg}\n\n"
yield ans
except Exception as f:
ans += "Произошла непредвиденная ошибка: " + str(f) + "\n"
finally:
if current_cmd:
message.delete_active(session)
yield str(ans)
elif 'Выход' in message.params and current_cmd:
message.delete_active(session)
yield {'msg': 'Успешно!', 'keyboard': [[], False]}
else:
if current_cmd is None:
message.add_setting(session, 'active', 'stt')
yield {'msg': 'Прикрепите аудио или напишите Выход',
'keyboard': [[[('Выход', 'negative')]], False]
}
def listen(self):
try:
with Microphone() as source:
recognizer = Recognizer()
recognizer.adjust_for_ambient_noise(source)
audio = recognizer.listen(source)
return recognizer.recognize_google(audio,
language=self.lang).lower()
except (UnknownValueError, RequestError):
return ''
def gettingWordsFromMic():
mic = Microphone()
recognizer = Recognizer()
print("Say something...")
print(Microphone.list_microphone_names())
with mic as source:
recognizer.adjust_for_ambient_noise(source)
audio = recognizer.listen(source)
input("Press a key to process")
print(recognizer.recognize_google(audio))
def speech_recog(file_name="input_sample/audio/introduction_ml.mp3",
duration=10):
convert_mp3_to_wav("input_sample/audio/introduction_ml.mp3")
result = 0
class_audio = AudioFile("input_sample/audio/introduction_ml.wav")
print(type(class_audio))
recongizer = Recognizer()
with class_audio as src_audio:
audio = recongizer.record(src_audio, duration=duration)
print(recongizer.recognize_google(audio))
def on_listen(recognizer: sr.Recognizer, audio: sr.AudioData):
l = len(audio.frame_data) / audio.sample_rate / audio.sample_width
e, tr = energy(recognizer, audio)
print(f'+{l:6.2f}s {e:6.2f}; [{recognizer.energy_threshold:.2f}]: ', end='')
try:
t = time()
text = recognizer.recognize_google(audio, language='ru-RU').lower()
except sr.UnknownValueError as e:
print()
else:
dt = time() - t
print(f'Вы сказали ({dt:5.2f}s): {text}')
def start(data):
if data.endswith('.wav'):
data = converter(data)
else:
print("Need Conversion")
r = Recognizer()
with AudioFile(data) as source:
audio = r.listen(source)
print("Status: Working\r", end="")
query = r.recognize_google(audio)
file = open("{}.txt".format(data.split('.')[0]), 'w')
file.write(query)
file.close()
print(query)
def audio_to_text(message_input):
# initialise the recognizer
r = Recognizer()
# Use the sysdefault microphone
for i, microphone_name in enumerate(Microphone.list_microphone_names()):
if microphone_name == "sysdefault":
micro = Microphone(device_index=i)
with micro as source:
# Extract the audio and convert it to text
audio = r.listen(source)
# recognize speech using Google Speech Recognition and add it to the text input area
try:
message_input.setText(r.recognize_google(audio))
except UnknownValueError:
message_input.setText('The audio was not understood')
class Listener:
def __init__(self) -> None:
super(Listener, self).__init__()
self._recognizer = Recognizer()
self._recognizer.energy_threshold = 4000
def listen(self) -> None:
"""Listens while the microphone is open and turns the audio into readable text."""
with Microphone() as source:
print('listening...')
self._recognizer.adjust_for_ambient_noise(source)
audio_listened = self._recognizer.listen(source)
text_listened: Any = self._recognizer.recognize_google(
audio_listened, )
self.text: str = text_listened.lower()
def get_audio():
rObject = Recognizer()
audio = ''
with Microphone() as source:
print("Speak...")
# recording the audio using speech recognition
audio = rObject.listen(source, phrase_time_limit=5)
print("Stop.") # limit 5 secs
try:
text = rObject.recognize_google(audio, language='en-US')
print("You: ", text)
return text
except:
chatbot_speaks("Could not understand your audio, Please try again !")
return 0
def myCommand(self):
#listens for commands
r = Recognizer()
with Microphone() as source:
print('Say something...')
r.pause_threshold = 1
r.adjust_for_ambient_noise(source, duration=1)
audio = r.listen(source)
try:
command = r.recognize_google(audio).lower()
print('You said: ' + command + '\n')
# loop back to continue to listen for commands if unrecognizable speech is received
except UnknownValueError:
print('....')
command = speech.myCommand(self)
return command
class VoiceService:
def __init__(self):
self.client = Recognizer()
def capture_audio(self):
with Microphone() as mic:
audio = self.client.listen(mic)
return audio
def get_transcript(self, audio, show_all=False):
return self.client.recognize_google(audio, show_all=show_all)
def capture_and_transcribe_audio(self):
audio = self.capture_audio()
#print(type(audio)) #> <class 'speech_recognition.AudioData'>
transcript = self.get_transcript(audio, show_all=False)
return transcript.upper()
class Listener(object):
def __init__(self):
self.__recognizer = Recognizer()
def __get_audio_from_file(self, filename, timeout=None, on_listen=None):
with AudioFile(filename) as source:
return self.__get_audio_from_source(source, timeout, on_listen)
def __get_audio_from_microphone(self, timeout=None, on_listen=None):
with Microphone() as source:
self.__recognizer.adjust_for_ambient_noise(source, duration=1.0)
return self.__get_audio_from_source(source, timeout, on_listen)
def __get_audio_from_source(self, source, timeout=None, on_listen=None):
if callable(on_listen): on_listen()
return self.__recognizer.listen(source, timeout=timeout)
def __recognize(self, audio, language, on_recognition=None):
if callable(on_recognition): on_recognition()
return self.__recognizer.recognize_google(audio, language=language)
def get_text_from_audio(self,
filename,
language="en-us",
timeout=None,
on_listen=None,
on_recognition=None):
try:
audio = self.__get_audio_from_file(filename, timeout, on_listen)
return self.__recognize(audio, language, on_recognition)
except FileNotFoundError as error:
raise error
except:
return str()
def speech_to_text(self,
language="en-us",
timeout=None,
on_listen=None,
on_recognition=None):
try:
audio = self.__get_audio_from_microphone(timeout, on_listen)
return self.__recognize(audio, language, on_recognition)
except:
return str()
def listening_callback(recognizer: sr.Recognizer, audio) -> None:
"""
Callback to be called when a phrase was recorded
:param recognizer: sr.Recognizer to be used
:param audio: phrase as audio
"""
try:
# try to use Google speech recognition to extract the text.
# the Google one works best (out of the free ones)
txt = recognizer.recognize_google(audio,
language=config.recognizer_lang)
log("text is: {0}".format(txt), remote=False)
send_text_to_server(txt)
except sr.UnknownValueError: # happens if the detected phrase is empty (= silence) or cannot be detected
log("Unknown value", level=logging.ERROR, remote=False)
except sr.RequestError:
log("Request error", level=logging.ERROR)
def read_mic_input(r: sr.Recognizer, mic: sr.Microphone) -> Dict:
"""Use r to transcribe speech recorded from mic, and return a dictionary
containing three keys:
"success": boolean value indicating whether or not the recognizer's speech
transcription was successful or not
"error": 'None' if no error occured, or a string containing the error
message if an error occured.
"transcription": 'None' if speech could not be transcribed, or a string
containing the transcribed text.
"""
with mic as source:
# adjust the recognizer sensitivity to account for ambient noise
r.adjust_for_ambient_noise(source, duration=0.3)
# Record voice input from microhpone
audio = r.listen(source)
# intialize the response dictionary to be returned
response = {"success": True, "error": None, "transcription": None}
# Attempt to recognize speech in the recording
try:
response["transcription"] = r.recognize_google(audio).lower()
# clean up the transcription of coordinates and measurements
response["transcription"] = response["transcription"].replace("-", " ")
response["transcription"] = response["transcription"].replace("/", " ")
response["transcription"] = response["transcription"].replace(
"\\", " ")
response["transcription"] = response["transcription"].replace(
" 00", " 0 0")
# Update response object if a RequestError or UnknownValueError exception is
# caught
except sr.RequestError:
# API was unreachable or unresponsive
response["success"] = False
response["error"] = "Error occurred with the API request."
except sr.UnknownValueError:
# speech could not be transcribed
response["error"] = "Unable to recognize speech."
return response
def listen():
r = Recognizer() # less writing
with Microphone() as source: # using microphone to detect audio
r.adjust_for_ambient_noise(
source, duration=0.5) #adjust for ambiet sounds for 1 second
audio = r.listen(source) # listen for audio
data = '' # set data ad nothing
try: # in case of errors
data = r.recognize_google(
audio) # recognize with google's speech recognition
print('You said: ' + data) # write what was heard
except UnknownValueError: # unknown audio
print('I didn\'t get that') # when google api didn't understand audio
data = 'None' # return none
except RequestError as e: # request error
print('Api or connection is not working.\n The error is {0}'.format(
e)) # when connection or Api offline
data = 'Broken' # return broken
return data # return recognized audio as string
class RecognizerOfAudioSource(AudioSource):
"""Represent recogniser of audio source."""
def __init__(self) -> None:
self._recogniser = Recognizer()
def __enter__(self) -> Recognizer:
return self._recogniser.__enter__()
def __exit__(self, exc_type, exc_val, exc_tb) -> None:
return self._recogniser.__exit__(exc_type, exc_val, exc_tb)
@property
def pause(self) -> int:
"""Return seconds of non-speaking audio before a phrase is considered complete"""
return self._recogniser.pause_threshold
@pause.setter
def pause(self, seconds: int) -> None:
if not isinstance(seconds, int):
raise TypeError('Seconds should be <int> data type.')
def configure_noise(self, source: Microphone, duration: int = 1) -> None:
self._recogniser.adjust_for_ambient_noise(source, duration)
def listen(
self,
source: Microphone,
timeout: int = None,
phrase_time_limit: int = None,
config: str= None
) -> AudioData:
return self._recogniser.listen(source, timeout, phrase_time_limit, config)
def recognise_google(
self,
audio_data: AudioData,
key: int = None,
language: str = "en-US",
show_all: bool = False
) -> str:
return self._recogniser.recognize_google(audio_data, key, language, show_all).lower()
def record_and_recognise(r: sr.Recognizer) -> str:
with sr.Microphone() as source:
print("Слушаю..")
r.pause_threshold = 1
r.adjust_for_ambient_noise(source, duration=1)
audio = r.listen(source, phrase_time_limit=5)
try:
recognized = r.recognize_google(audio, language='ru-RU')
if type(recognized) == list:
recognized_string = " ".join(recognized)
return recognized_string
else:
return recognized
except sr.UnknownValueError:
print(
"Unknown sounds came from your mouth. Or perhaps we can't hear you."
)
except sr.RequestError as e:
print("Sound recognition exception; {0}".format(e))
async def process_audio(chat_id, msg):
await bot.download_file(msg['voice']['file_id'], "./dest.ogg")
filename = "dest.ogg"
dest = "dest.flac"
r = Recognizer()
sound = AudioSegment.from_ogg(filename)
os.unlink(filename)
sound.export(dest, format="flac")
with AudioFile(dest) as source:
# listen for the data (load audio to memory)
audio_data = r.record(source)
# recognize (convert from speech to text)
try:
text = r.recognize_google(audio_data)
print(f"VOICE LOG - {msg['from']['first_name']}: {text}")
await process_result(chat_id, text)
except UnknownValueError:
await bot.sendMessage(chat_id, 'This audio is too short or corrupted, retry!')
pass
try:
os.unlink(dest)
except PermissionError:
pass
def SpeechToText():
try:
rec = Recognizer()
with AudioFile(WAVE_OUTPUT_FILENAME) as AudioSrc:
content = rec.record(AudioSrc)
text = rec.recognize_google(
audio_data=content,
language='ko-KR',
)
mainText = str(text)
print('main-> ', mainText)
return mainText
# for sW in START_WORDS:
# if mainText.startswith(sW):
# return mainText.split(sW)[1]
except (UnknownValueError):
print("SPEECH ERROR!")
return False
def solve(self):
try:
sleep(3)
# Scrolling Down the Page
pyautogui.scroll(-1000)
# Locating and Clicking Captcha button on Page
cap = pyautogui.locateCenterOnScreen("files/captcha.png")
pyautogui.click(cap)
try:
sleep(3)
# Locating and Clicking Headphones button on Page
voi = pyautogui.locateCenterOnScreen("files/voice.png")
pyautogui.click(voi)
except:
try:
# Move out mouse from the voice button
pyautogui.moveTo(200, 200)
# Locate Voice2 Button that is little Gray
pyautogui.locateCenterOnScreen("files/voice2.png")
return self.error
except:
return self.done
except:
try:
sleep(2)
# Move out mouse from the voice button
pyautogui.moveTo(200, 200)
# Locate Voice2 Button that is little Gray
pyautogui.locateCenterOnScreen("files/voice2.png")
self.error[1] = "Captcha"
return self.error
except:
return self.done
sleep(2)
try:
# Locate and Right Click on the Download Button
down = pyautogui.locateCenterOnScreen("files/down.png")
pyautogui.rightClick(down)
sleep(1)
# Press down button 5 times
pyautogui.press(['down'] * 5)
sleep(1)
# Press Enter (Cursor will be Copy Link Address)
pyautogui.press('enter')
# Download and Save that audio file from Link copied from above code
with open('files/audio.mp3', 'wb') as file:
r = ge(paste())
file.write(r.content)
sleep(2)
if path.exists('files/audio.mp3'):
# Convert that mp3 file into wav using ffpmeg
call([
'files/ffmpeg.exe', '-i', 'files/audio.mp3', '-y',
'files/audio.wav'
])
sleep(2)
AUDIO_FILE = 'files/audio.wav'
# Code to Send That Audio File to Google and Recognize The Audio
r = Recognizer()
with AudioFile(AUDIO_FILE) as source:
audio = r.record(source)
try:
# Get the Recognized Text
capSolved = r.recognize_google(audio)
# Go to the text field and write it there
pyautogui.hotkey('shift', 'tab')
pyautogui.typewrite(capSolved)
pyautogui.press('enter')
sleep(5)
return self.done
except UnknownValueError as e:
return self.error
except RequestError as e:
return self.error
except:
self.error[1] = "Captcha"
return self.error
from speech_recognition import Microphone, RequestError, Recognizer, UnknownValueError
dir(sr)
import pyaudio
import webbrowser as wb
r1 = Recognizer() # this is the recognizer class from speechRecognition
r2 = Recognizer()
r3 = Recognizer()
with sr.Microphone as source:
print('[search Google: search YouTube]')
print('Speak Now!!')
audio = r3.listen(source)
if 'video' in r1.recognize_google(audio):
r1 = Recognizer()
url = 'https://www.youtube.com/results?search_query='
with sr.Microphone() as source:
print('search your query')
audio = r1.listen(source)
try:
get = r1.recognize_google(audio)
print(get)
wb.get().open_new(url + get)
except UnknownValueError:
print('error ')
except RequestError as e:
print('failed'.format(e))
from speech_recognition import Microphone, Recognizer
recog=Recognizer()
mic=Microphone()
with mic:
print("talk")
audio=recog.listen(mic)
recognize=recog.recognize_google(audio)
print(recognize)