def main():
parser = configargparse.ArgParser(
default_config_files=CONFIG_FILES,
description="Act on voice commands using Google's speech recognition")
parser.add_argument('-I', '--input-device', default='default',
help='Name of the audio input device')
parser.add_argument('-O', '--output-device', default='default',
help='Name of the audio output device')
parser.add_argument('-T', '--trigger', default='gpio',
help='Trigger to use {\'clap\', \'gpio\'}')
parser.add_argument('--cloud-speech', action='store_true',
help='Use the Cloud Speech API instead of the Assistant API')
parser.add_argument('-L', '--language', default='en-US',
help='Language code to use for speech (default: en-US)')
parser.add_argument('-l', '--led-fifo', default='/tmp/status-led',
help='Status led control fifo')
parser.add_argument('-p', '--pid-file',
help='File containing our process id for monitoring')
parser.add_argument('--audio-logging', action='store_true',
help='Log all requests and responses to WAV files in /tmp')
parser.add_argument('--assistant-secrets',
default=os.path.expanduser('~/assistant.json'),
help='Path to client secrets for the Assistant API')
parser.add_argument('--cloud-speech-secrets',
default=os.path.expanduser('~/cloud_speech.json'),
help='Path to service account credentials for the '
'Cloud Speech API')
parser.add_argument('--trigger-sound', default=None,
help='Sound when trigger is activated (WAV format)')
args = parser.parse_args()
create_pid_file(args.pid_file)
i18n.set_language_code(args.language, gettext_install=True)
player = audio.Player(args.output_device)
if args.cloud_speech:
credentials_file = os.path.expanduser(args.cloud_speech_secrets)
if not os.path.exists(credentials_file) and os.path.exists(OLD_SERVICE_CREDENTIALS):
credentials_file = OLD_SERVICE_CREDENTIALS
recognizer = speech.CloudSpeechRequest(credentials_file)
else:
credentials = try_to_get_credentials(
os.path.expanduser(args.assistant_secrets))
recognizer = speech.AssistantSpeechRequest(credentials)
recorder = audio.Recorder(
input_device=args.input_device, channels=1,
bytes_per_sample=speech.AUDIO_SAMPLE_SIZE,
sample_rate_hz=speech.AUDIO_SAMPLE_RATE_HZ)
with recorder:
do_recognition(args, recorder, recognizer, player)
def record_button(self):
"""
Records high quality audio files on the artist side.
:return:
"""
# Toggle recording
rec_state = self.app.get_own_state()['recording']
self.app.get_own_state()['recording'] = not rec_state
if self.app.get_own_state()['recording']:
# Update state
self.app.chat_client.send_sync(constants.SYNC_START_RECORDING)
# GUI update
self.ids.record_button.source = SessionScreen.stop_theme
# Start the progress effect
progress_thread = threading.Thread(target=self.record_progress)
progress_thread.start()
filename = self.app.config.get_file_name(self.app.session_name,
datetime.now().strftime(constants.DATETIME_HQ))
head, tail = os.path.split(filename)
print "Creating: " + tail
# Make sure folders exist
if not os.path.exists(head):
os.makedirs(head)
# Makes a Recorder with the desired filename
self.app.recorder = audio.Recorder(filename)
# Add available audio file to state list
self.app.add_audio_file(filename)
# Starts writing to an audio file, including to disk
self.app.recorder.start() # Starts recording
print "Recording..."
else:
# Update state
self.app.chat_client.send_sync(constants.SYNC_STOP_RECORDING)
# GUI update
self.ids.record_button.source = SessionScreen.record_red
# Closes recording threads
self.app.recorder.stop()
self.app.phone.stop_start_recording(datetime.now().strftime(constants.DATETIME_LQ))
self.add_clip() # adds to gui sidebar
# Send a sync message for when a clip is available
available_filename = self.app.get_latest_audio_file()
_, tail = os.path.split(available_filename)
self.app.chat_client.send_sync(constants.SYNC_FILE_AVAILABLE,
filename=tail,
length=audio.get_length(available_filename))
new_colors = [colorsys.hsv_to_rgb(h, 1.0, 1.0) for h in new_hues]
old_cols = copy.deepcopy(new_colors)
for i in xrange(len(new_colors)):
if i not in [i1, i2]:
new_colors[i] = (0.0, 0.0, 0.0)
vis.state = new_colors
if not disp:
print new_colors
vis.write_state()
return scale, i1, i2, old_cols
if __name__ == '__main__':
disp = True
vis = comms.Visualizer(debug=False)
rec = audio.Recorder()
fps = 106.0
spf = 1.0 / fps
i1 = 0
i2 = 7
old_cols = [(0.0, 0.0, 0.0), (0.0, 0.0, 0.0), (0.0, 0.0, 0.0),
(0.0, 0.0, 0.0), (0.0, 0.0, 0.0), (0.0, 0.0, 0.0),
(0.0, 0.0, 0.0), (0.0, 0.0, 0.0)]
if disp:
display = ui.Display()
while True:
buff = ''
sys.stdin.flush()
while len(buff) < 100:
buff += sys.stdin.read(1)
#if len(buff)>100:
def main():
parser = configargparse.ArgParser(
default_config_files=CONFIG_FILES,
description="Act on voice commands using Google's speech recognition")
parser.add_argument('-I',
'--input-device',
default='default',
help='Name of the audio input device')
parser.add_argument('-O',
'--output-device',
default='default',
help='Name of the audio output device')
parser.add_argument('-T',
'--trigger',
default='gpio',
choices=['clap', 'gpio', 'ok-google'],
help='Trigger to use')
parser.add_argument(
'--cloud-speech',
action='store_true',
help='Use the Cloud Speech API instead of the Assistant API')
parser.add_argument(
'-L',
'--language',
default='en-US',
help='Language code to use for speech (default: en-US)')
parser.add_argument('-l',
'--led-fifo',
default='/tmp/status-led',
help='Status led control fifo')
parser.add_argument('-p',
'--pid-file',
help='File containing our process id for monitoring')
parser.add_argument(
'--audio-logging',
action='store_true',
help='Log all requests and responses to WAV files in /tmp')
parser.add_argument('--assistant-always-responds',
action='store_true',
help='Play Assistant responses for local actions.'
' You should make sure that you have IFTTT applets for'
' your actions to get the correct response, and also'
' that your actions do not call say().')
parser.add_argument('--assistant-secrets',
default=os.path.expanduser('~/assistant.json'),
help='Path to client secrets for the Assistant API')
parser.add_argument('--cloud-speech-secrets',
default=os.path.expanduser('~/cloud_speech.json'),
help='Path to service account credentials for the '
'Cloud Speech API')
parser.add_argument('--trigger-sound',
default=None,
help='Sound when trigger is activated (WAV format)')
args = parser.parse_args()
create_pid_file(args.pid_file)
i18n.set_language_code(args.language, gettext_install=True)
player = audio.Player(args.output_device)
if args.cloud_speech:
credentials_file = os.path.expanduser(args.cloud_speech_secrets)
if not os.path.exists(credentials_file) and os.path.exists(
OLD_SERVICE_CREDENTIALS):
credentials_file = OLD_SERVICE_CREDENTIALS
recognizer = speech.CloudSpeechRequest(credentials_file)
else:
credentials = try_to_get_credentials(
os.path.expanduser(args.assistant_secrets))
recognizer = speech.AssistantSpeechRequest(credentials)
status_ui = StatusUi(player, args.led_fifo, args.trigger_sound)
# The ok-google trigger is handled with the Assistant Library, so we need
# to catch this case early.
if args.trigger == 'ok-google':
if args.cloud_speech:
print('trigger=ok-google only works with the Assistant, not with '
'the Cloud Speech API.')
sys.exit(1)
do_assistant_library(args, credentials, player, status_ui)
else:
recorder = audio.Recorder(input_device=args.input_device,
channels=1,
bytes_per_sample=speech.AUDIO_SAMPLE_SIZE,
sample_rate_hz=speech.AUDIO_SAMPLE_RATE_HZ)
with recorder:
do_recognition(args, recorder, recognizer, player, status_ui)
if var_energy == 0.0:
return False
#if not disp:
# print var_energy,var_e/var_energy,var_prev/var_energy
if ((var_prev / var_energy) < thresh) and ((var_e / var_energy) > thresh):
if not disp:
print True
return True
else:
return False
if __name__ == '__main__':
disp = True
vis = comms.Visualizer(debug=False)
rec = audio.Recorder(sample_size=1000)
fps = 106.0
spf = 1.0 / fps
bar_i = 1
direc = 1
n_energies = 30
energies = [0.0 for i in xrange(n_energies)]
if disp:
display = ui.Display()
while True:
time_0 = time.time()
buff = ''
sys.stdin.flush()
while len(buff) < 1000:
buff += sys.stdin.read(1)
#if len(buff)>100:
def main():
global mqttclient, Dplayer
parser = configargparse.ArgParser(
default_config_files=CONFIG_FILES,
description="Act on voice commands using Google's speech recognition")
parser.add_argument('-I',
'--input-device',
default='default',
help='Name of the audio input device')
parser.add_argument('-O',
'--output-device',
default='default',
help='Name of the audio output device')
parser.add_argument('-T',
'--trigger',
default='gpio',
help='Trigger to use {\'clap\', \'gpio\'}')
parser.add_argument(
'--cloud-speech',
action='store_true',
help='Use the Cloud Speech API instead of the Assistant API')
parser.add_argument(
'-L',
'--language',
default='en-US',
help='Language code to use for speech (default: en-US)')
parser.add_argument('-l',
'--led-fifo',
default='/tmp/status-led',
help='Status led control fifo')
parser.add_argument('-p',
'--pid-file',
default=PID_FILE,
help='File containing our process id for monitoring')
parser.add_argument(
'--audio-logging',
action='store_true',
help='Log all requests and responses to WAV files in /tmp')
parser.add_argument('--assistant-secrets',
help='Path to client secrets for the Assistant API')
parser.add_argument('--cloud-speech-secrets',
help='Path to service account credentials for the '
'Cloud Speech API')
args = parser.parse_args()
create_pid_file(args.pid_file)
i18n.set_language_code(args.language, gettext_install=True)
player = audio.Player(args.output_device)
Dplayer = player
action.VolumeControl(None, 100).run(
"max volume") # start loud at the start (which is not very loud)
blipsay(piblips + "Booting up 1 2 3 4 5")
if args.cloud_speech:
credentials_file = os.path.expanduser(args.cloud_speech_secrets)
if not os.path.exists(credentials_file) and os.path.exists(
OLD_SERVICE_CREDENTIALS):
credentials_file = OLD_SERVICE_CREDENTIALS
recognizer = speech.CloudSpeechRequest(credentials_file)
else:
credentials = try_to_get_credentials(
os.path.expanduser(args.assistant_secrets))
recognizer = speech.AssistantSpeechRequest(credentials)
recorder = audio.Recorder(input_device=args.input_device,
channels=1,
bytes_per_sample=speech.AUDIO_SAMPLE_SIZE,
sample_rate_hz=speech.AUDIO_SAMPLE_RATE_HZ)
# most is encapsulated here (then look for mqttclient.loop())
if mqttbroker:
mqttclient = paho.mqtt.client.Client(mqttbroker) # global object
def on_connect(client, userdata, flags, rc):
print("on_connect", client, userdata, flags, rc)
blipsay(piblips + "M Q T T Connected")
def on_disconnect(client, userdata, rc):
print("on_disconnect", client, userdata, rc)
blipsay(piblips + "M Q T T Disconnected")
def on_publish(client, userdata, mid):
print("on_publish", client, userdata, mid)
def on_message(client, userdata, message):
print("on_message", client, message, message.topic,
message.payload)
heardhistory.append(message.payload.decode())
blipsay(message.payload)
mqttclient.on_connect = on_connect
mqttclient.on_publish = on_publish
mqttclient.on_disconnect = on_disconnect
mqttclient.on_message = on_message
print("made and attempting to connect", mqttclient)
mqttclient.connect(mqttbroker)
mqttclient.subscribe(mqtttohear)
# this function enters an infinite while loop
with recorder:
do_recognition(args, recorder, recognizer, player)
