UnicodeDecodeError when Transcribing Live Audio using Sphinx in Python

Question

I am trying to transcribe a live playing audio, but I am getting the following error when decoding,

UnicodeDecodeError: 'utf-8' codec can't decode byte 0xff in position 0: invalid start byte

import pyaudio
import requests
from pocketsphinx import LiveSpeech


def speech_to_text(audio_stream):
    speech = LiveSpeech(
        verbose=False,
        sampling_rate=16000,
        buffer_size=2048,
        no_search=False,
    )

    for chunk in audio_stream.iter_content(chunk_size=1024):
        for phrase in speech.decode(bytes(chunk)):
            yield str(phrase)


def main():
    url = "AUDIO_URL"
    audio_stream = requests.get(url, stream=True)

    audio_format = pyaudio.paInt16
    sample_rate = 16000
    audio_chunk_size = 1024

    p = pyaudio.PyAudio()
    stream = p.open(format=audio_format, channels=1, rate=sample_rate, input=True, frames_per_buffer=audio_chunk_size)

    print("Listening to audio stream...")

    for phrase in speech_to_text(audio_stream):
        print("Recognized text:", phrase)

    stream.stop_stream()
    stream.close()
    p.terminate()


if __name__ == "__main__":
    main()

The error occurs when I attempt to process the binary audio data received from the URL. Since this is binary audio data, I understand that decoding it as UTF-8 doesn't make sense, but I'm unsure about how to handle it correctly.

So what I am doing wrong here? Can anybody help me? Thanks in advance.