ivanistheone/README.md

## README.md

      
    Raw
  

              README.md
            
          
    Speech recognition demo by ggulati

Original https://gist.github.com/GGulati/1ebaeaaa7f7408647fef
via https://ggulati.wordpress.com/2016/02/24/coding-jarvis-in-python-3-in-2016/
via https://news.ycombinator.com/item?id=11172727
I did some modifications to try to get it to work on Mac OS.
Install

These are the instructions for Mac OS 10.10.x
brew install portaudio flac swig
pip install -r requirements.txt

Note installing pyobjc pulls in a kitchen sink full of stuff, so you'll have
to go get a coffee.
Super fix

Monkey patch the file lib/python2.7/site-packages/pyttsx/drivers/nsss.py
by adding the line
from objc import super

at the top of the file.
Still no cigar

Apparently pyttsx isn't maintained anymore, so the program hangs:
RapidWareTech/pyttsx#26

  
## jarvis.py
from __future__ import print_function
import speech_recognition
import pyttsx

speech_engine = pyttsx.init('nsss')
speech_engine.setProperty('rate', 150)

recognizer = speech_recognition.Recognizer()
microphone = speech_recognition.Microphone()

def listen():
    with microphone as source:
        recognizer.adjust_for_ambient_noise(source)
        audio = recognizer.listen(source)
        print("Got it! Now to recognize it...")

    try:
        # recognizer.recognize_sphinx(audio)   # == not very good out of the box
        value = recognizer.recognize_google(audio)
        return value
    except speech_recognition.UnknownValueError:
        print("Could not understand audio")
    except speech_recognition.RequestError as e:
        print("Recog Error; {0}".format(e))
    return ""

speech_engine.say('Say something!')
# speech_engine.runAndWait()              # hangs here...
said = listen()
print(said)
# speech_engine.say('I heard you say ' + said)
# speech_engine.runAndWait()              # hangs here...

## requirements.txt
altgraph==0.12
macholib==1.7
modulegraph==0.12.1
py2app==0.9
PyAudio==0.2.9
pyobjc==3.0.4
pyttsx==1.1
SpeechRecognition==3.3.0
pocketsphinx==0.0.9
	from __future__ import print_function
	import speech_recognition
	import pyttsx

	speech_engine = pyttsx.init('nsss')
	speech_engine.setProperty('rate', 150)

	recognizer = speech_recognition.Recognizer()
	microphone = speech_recognition.Microphone()

	def listen():
	with microphone as source:
	recognizer.adjust_for_ambient_noise(source)
	audio = recognizer.listen(source)
	print("Got it! Now to recognize it...")

	try:
	# recognizer.recognize_sphinx(audio) # == not very good out of the box
	value = recognizer.recognize_google(audio)
	return value
	except speech_recognition.UnknownValueError:
	print("Could not understand audio")
	except speech_recognition.RequestError as e:
	print("Recog Error; {0}".format(e))
	return ""

	speech_engine.say('Say something!')
	# speech_engine.runAndWait() # hangs here...
	said = listen()
	print(said)
	# speech_engine.say('I heard you say ' + said)
	# speech_engine.runAndWait() # hangs here...
	altgraph==0.12
	macholib==1.7
	modulegraph==0.12.1
	py2app==0.9
	PyAudio==0.2.9
	pyobjc==3.0.4
	pyttsx==1.1
	SpeechRecognition==3.3.0
	pocketsphinx==0.0.9