speech.py

#!/usr/bin/env python3
# Requires PyAudio and PySpeech.

import pyaudio 
import speech_recognition as sr


# Record Audio
r = sr.Recognizer()

with sr.Microphone() as source:
    print("Say something!")
    r.adjust_for_ambient_noise(source, duration = 1)
    audio = r.listen(source)
 
# Speech recognition using Google Speech Recognition
try:
    # for testing purposes, we're just using the default API key
    # to use another API key, use `r.recognize_google(audio, key="GOOGLE_SPEECH_RECOGNITION_API_KEY")`
    # instead of `r.recognize_google(audio)`
    print("You said: " + r.recognize_google(audio))
except sr.UnknownValueError:
    print("Google Speech Recognition could not understand audio")
except sr.RequestError as e:
    print("Could not request results from Google Speech Recognition service; {0}".format(e))