Python3 SPEECH to TEXT - python-3.x

Question

I am a 12 year old kid and am a beginner at programming. It would be great if someone could help be. below I have the code for my speech recognition application, I am on MAC OS CATALINA and the same error keeps on coming up, It prints out "say something", and then once I say something nothing happens and it stays frozen, once I stop the code running I get this error.
import speech_recognition as sr
r = sr.Recognizer()
with sr.Microphone() as source:
print("SAY SOMETHING")
audio = r.listen(source)
print("THANK YOU")
try:
print("TEXT: "+r.recognize_google(audio))
except:
print("SORRY I DONT KNOW WHAT YOU MEAN")
This is the error I get when I stop the code, once it is paused at SAY SOMETHING for a long time.
Traceback (most recent call last):
File "/Users/anishnagariya/PycharmProjects/HelloWorld/Tester.py", line 8, in <module>
print("THANK YOU")
File "/Users/anishnagariya/PycharmProjects/AI/HelloWorld/lib/python3.7/site-packages/speech_recognition/__init__.py", line 620, in listen
buffer = source.stream.read(source.CHUNK)
File "/Users/anishnagariya/PycharmProjects/AI/HelloWorld/lib/python3.7/site-packages/speech_recognition/__init__.py", line 161, in read
return self.pyaudio_stream.read(size, exception_on_overflow=False)
File "/Users/anishnagariya/PycharmProjects/AI/HelloWorld/lib/python3.7/site-packages/pyaudio.py", line 608, in read
return pa.read_stream(self._stream, num_frames, exception_on_overflow)
KeyboardInterrupt
Process finished with exit code 1

Answer 1

I would advise going through the documentation for the speech recognition module you're using:
https://github.com/Uberi/speech_recognition/blob/master/reference/library-reference.rst
Check this part specifically:
https://github.com/Uberi/speech_recognition/blob/master/reference/library-reference.rst#recognizer_instanceenergy_threshold--300---type-float
The energy threshold is described as such:
Represents the energy level threshold for sounds. Values below this threshold are considered silence, and values above this threshold are considered speech. Can be changed.
You can set the threshold like this:
r.energy_threshold = 4000
Or you can do this which will adjust the threshold dynamically based on sound currently from the environment.
r.dynamic_energy_threshold = True
Also ensure your microphone is working properly
Edit:
https://github.com/Uberi/speech_recognition/blob/master/reference/library-reference.rst#recognizer_instancelistensource-audiosource-timeout-unionfloat-none--none-phrase_time_limit-unionfloat-none--none-snowboy_configuration-uniontuplestr-iterablestr-none--none---audiodata
Set a timeout &/or pause duration for the listen call.
import speech_recognition as sr
r = sr.Recognizer()
while True:
with sr.Microphone() as source:
print("SAY SOMETHING")
try:
audio = r.listen(source, timeout=3)
print("THANK YOU")
break
except sr.WaitTimeoutError:
print("timed out")
try:
print("TEXT: "+r.recognize_google(audio))
except:
print("SORRY I DONT KNOW WHAT YOU MEAN")

Answer 2

Depending on your microphone quality you will probably need to set some thresholds:
import speech_recognition as sr
r = sr.Recognizer()
r.energy_threshold = 1000
r.pause_threshold = 0.5
with sr.Microphone() as source:
print("SAY SOMETHING")
audio = r.listen(source)
print("THANK YOU")
try:
print("TEXT: " + r.recognize_google(audio))
except sr.UnknownValueError:
print("SORRY I DONT KNOW WHAT YOU MEAN")
It is also not good practice to have a bare except clause sr.UnknownValueError is the standard error for unknown speech in the speech_recognition library