def do_recognition(args, recorder, recognizer, player, status_ui):
"""Configure and run the recognizer."""
say = aiy.audio.say
actor = action.make_actor(say)
if args.cloud_speech:
action.add_commands_just_for_cloud_speech_api(actor, say)
recognizer.add_phrases(actor)
recognizer.set_audio_logging_enabled(args.audio_logging)
if args.trigger == 'gpio':
import triggers.gpio
triggerer = triggers.gpio.GpioTrigger(channel=23)
msg = 'Press the button on GPIO 23'
elif args.trigger == 'clap':
import triggers.clap
triggerer = triggers.clap.ClapTrigger(recorder)
msg = 'Clap your hands'
else:
logger.error("Unknown trigger '%s'", args.trigger)
return
mic_recognizer = SyncMicRecognizer(
actor, recognizer, recorder, player, say, triggerer, status_ui,
args.assistant_always_responds)
with mic_recognizer:
if sys.stdout.isatty():
print(msg + ' then speak, or press Ctrl+C to quit...')
# wait for KeyboardInterrupt
while True:
time.sleep(1)
def do_recognition(args, recorder, recognizer, player, status_ui):
"""Configure and run the recognizer."""
say = aiy.audio.say
actor = action.make_actor(say)
if args.cloud_speech:
action.add_commands_just_for_cloud_speech_api(actor, say)
recognizer.add_phrases(actor)
recognizer.set_audio_logging_enabled(args.audio_logging)
if args.trigger == 'gpio':
import triggers.gpio
triggerer = triggers.gpio.GpioTrigger(channel=23)
msg = 'Press the button on GPIO 23'
elif args.trigger == 'clap':
import triggers.clap
triggerer = triggers.clap.ClapTrigger(recorder)
msg = 'Clap your hands'
else:
logger.error("Unknown trigger '%s'", args.trigger)
return
mic_recognizer = SyncMicRecognizer(actor, recognizer, recorder, player,
say, triggerer, status_ui,
args.assistant_always_responds)
with mic_recognizer:
if sys.stdout.isatty():
print(msg + ' then speak, or press Ctrl+C to quit...')
# wait for KeyboardInterrupt
while True:
time.sleep(1)
def do_recognition(args, recorder, recognizer, player):
"""Configure and run the recognizer."""
global Dplayer, Dsay, Drecorder, Drecognizer, Dtriggerer
say = tts.create_say(player)
actor = action.make_actor(say)
Dplayer = player
Dsay = say
Drecorder = recorder
Drecognizer = recognizer
if args.cloud_speech:
action.add_commands_just_for_cloud_speech_api(actor, say)
recognizer.add_phrases(actor)
recognizer.set_audio_logging_enabled(args.audio_logging)
if args.trigger == 'gpio':
import triggers.gpio
triggerer = triggers.gpio.GpioTrigger(channel=23)
msg = 'Press the button on GPIO 23'
elif args.trigger == 'clap':
import triggers.clap
triggerer = triggers.clap.ClapTrigger(recorder)
msg = 'Clap your hands'
else:
logger.error("Unknown trigger '%s'", args.trigger)
return
Dtriggerer = triggerer
mic_recognizer = SyncMicRecognizer(actor,
recognizer,
recorder,
player,
say,
triggerer,
led_fifo=args.led_fifo)
with mic_recognizer:
if sys.stdout.isatty():
print(msg + ' then speak, or press Ctrl+C to quit...')
# wait for KeyboardInterrupt
i = 0
while True:
i += 1
if mqttclient:
mqttclient.loop(
timeout=0.2
) # needs to be called regularly (otherwise would need its own thread)
else:
time.sleep(0.2)
if (i % 100) == 0:
logger.info("tick")
def do_assistant_library(args, recognizer, credentials, player, status_ui):
"""Run a recognizer using the Google Assistant Library.
The Google Assistant Library has direct access to the audio API, so this
Python code doesn't need to record audio.
"""
try:
from google.assistant.library import Assistant
from google.assistant.library.event import EventType
except ImportError:
print('''
ERROR: failed to import the Google Assistant Library. This is required for
"OK Google" hotwording, but is only available for Raspberry Pi 2/3. It can be
installed with:
env/bin/pip install google-assistant-library==0.0.2''')
sys.exit(1)
say = tts.create_say(player)
actor = action.make_actor(say)
action.add_commands_just_for_cloud_speech_api(actor, say)
recognizer.add_phrases(actor)
def process_event(event):
logging.info(event)
if event.type == EventType.ON_START_FINISHED:
status_ui.status('ready')
if sys.stdout.isatty():
print(
'Say "OK, Google" then speak, or press Ctrl+C to quit...')
elif event.type == EventType.ON_CONVERSATION_TURN_STARTED:
status_ui.status('listening')
elif event.type == EventType.ON_END_OF_UTTERANCE:
status_ui.status('thinking')
elif event.type == EventType.ON_RECOGNIZING_SPEECH_FINISHED and \
event.args and actor.can_handle(event.args['text']):
if not args.assistant_always_responds:
assistant.stop_conversation()
actor.handle(event.args['text'])
elif event.type == EventType.ON_CONVERSATION_TURN_FINISHED:
status_ui.status('ready')
elif event.type == EventType.ON_ASSISTANT_ERROR and \
event.args and event.args['is_fatal']:
sys.exit(1)
with Assistant(credentials) as assistant:
for event in assistant.start():
process_event(event)
def do_assistant_library(args, credentials, player, status_ui):
"""Run a recognizer using the Google Assistant Library.
The Google Assistant Library has direct access to the audio API, so this
Python code doesn't need to record audio.
"""
try:
from google.assistant.library import Assistant
from google.assistant.library.event import EventType
except ImportError:
print('''
ERROR: failed to import the Google Assistant Library. This is required for
"OK Google" hotwording, but is only available for Raspberry Pi 2/3. It can be
installed with:
env/bin/pip install google-assistant-library==0.0.2''')
sys.exit(1)
say = aiy.audio.say
actor = action.make_actor(say)
def process_event(event):
logging.info(event)
if event.type == EventType.ON_START_FINISHED:
status_ui.status('ready')
if sys.stdout.isatty():
print('Say "OK, Google" then speak, or press Ctrl+C to quit...')
elif event.type == EventType.ON_CONVERSATION_TURN_STARTED:
status_ui.status('listening')
elif event.type == EventType.ON_END_OF_UTTERANCE:
status_ui.status('thinking')
elif event.type == EventType.ON_RECOGNIZING_SPEECH_FINISHED and \
event.args and actor.can_handle(event.args['text']):
if not args.assistant_always_responds:
assistant.stop_conversation()
actor.handle(event.args['text'])
elif event.type == EventType.ON_CONVERSATION_TURN_FINISHED:
status_ui.status('ready')
elif event.type == EventType.ON_ASSISTANT_ERROR and \
event.args and event.args['is_fatal']:
sys.exit(1)
with Assistant(credentials) as assistant:
for event in assistant.start():
process_event(event)
def do_recognition(args, recorder, recognizer, player, credentials):
"""Configure and run the recognizer."""
say = tts.create_say(player)
actor = action.make_actor(say)
if args.cloud_speech:
action.add_commands_just_for_cloud_speech_api(actor, say)
recognizer.add_phrases(actor)
recognizer.set_audio_logging_enabled(args.audio_logging)
if args.trigger == 'gpio':
import triggers.gpio
triggerer = triggers.gpio.GpioTrigger(channel=23)
msg = 'Press the button on GPIO 23'
elif args.trigger == 'clap':
import triggers.clap
triggerer = triggers.clap.ClapTrigger(recorder)
msg = 'Clap your hands'
elif args.trigger == 'hotword':
import triggers.hotword
triggerer = triggers.hotword.HotwordTrigger(credentials)
msg = 'Say "Ok Google"'
else:
logger.error("Unknown trigger '%s'", args.trigger)
return
mic_recognizer = SyncMicRecognizer(actor,
recognizer,
recorder,
player,
say,
triggerer,
led_fifo=args.led_fifo)
with mic_recognizer:
if sys.stdout.isatty():
print(msg + ' then speak, or press Ctrl+C to quit...')
# wait for KeyboardInterrupt
while True:
time.sleep(1)
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',
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)
say = tts.create_say(player)
actor = action.make_actor(say)
listener = tj.Client(actor, say, player)
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)
