utils/audio.py

import soundfile as sf
import numpy as np

def load_audio_mono_16k(path: str):
    wav, sr = sf.read(path, dtype="float32", always_2d=False)
    if wav.ndim == 2:
        wav = wav.mean(axis=1)
    if sr != 16000:
        # lightweight resample
        import numpy as np
        import math
        ratio = 16000 / sr
        n = int(math.floor(len(wav) * ratio))
        x_old = np.linspace(0, 1, len(wav), endpoint=False)
        x_new = np.linspace(0, 1, n, endpoint=False)
        wav = np.interp(x_new, x_old, wav).astype("float32")
        sr = 16000
    return wav, sr