Wake word detection in porcupine

I'm trying to implement porcupine wakeword in python and have followed the instructions from here

I have the following code:

import pvporcupine

### Porcupine wakeword
handle = pvporcupine.create(keywords=['computer', 'jarvis'])

def get_next_audio_frame():
    pass

while True:
    keyword_index = handle.process(get_next_audio_frame())
    if keyword_index >= 0:
        # Insert detection event callback here
        print('Yes sir?')
        pass

but I receive the following error:

❯ python3 porcupine.py Traceback (most recent call last):   File "porcupine.py", line 10, in <module>
    keyword_index = handle.process(get_next_audio_frame())   File "/home/rupstar/Computer/lib/python3.8/site-packages/pvporcupine/porcupine.py", line 129, in process
    if len(pcm) != self.frame_length: TypeError: object of type 'NoneType' has no len()

This may not be perfect (it isn't), but it shows how I've created a personal voice assistant in python on WSL2 running Ubuntu 20.04 on a Windows 10 machine. The voice assistant responds to a wakeword (Jarvis or Computer) and then executes commands. Relevant to this post is how porcupine is invoked:

#!/usr/bin/env python3
#Porcupine wakeword includes
import struct
import pyaudio
import pvporcupine

porcupine = None
pa = None
audio_stream = None

try:
    porcupine = pvporcupine.create(keywords=["computer", "jarvis"])

    pa = pyaudio.PyAudio()

    audio_stream = pa.open(
                    rate=porcupine.sample_rate,
                    channels=1,
                    format=pyaudio.paInt16,
                    input=True,
                    frames_per_buffer=porcupine.frame_length)

    while True:
        pcm = audio_stream.read(porcupine.frame_length)
        pcm = struct.unpack_from("h" * porcupine.frame_length, pcm)

        keyword_index = porcupine.process(pcm)

        if keyword_index >= 0:
            print("Hotword Detected")
            speak("Computer online")