testsr.py

import speech_recognition as sr
r = sr.Recognizer()
with sr.Microphone() as source:
     r.adjust_for_ambient_noise(source)
     print("Say something!")
     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))