config.json

{
  "audio_config": {
    "format": "paInt16",
    "channels": 1,
    "rate": 16000,
    "chunk": 1024
  },
  "audio_fft_config": {
    "nfft": 256,
    "hop_length": 128,
    "amplitude": 32768
  },
  "background_mask_config": {
    "config_type": "Kirigami_background_mask",
    "heuristic_sampling_config": {
      "calibration_samples_frame": 10,
      "samples_per_frame": 100,
      "sample_count": 130
    },
    "mask_filter_config":{
      "n_grad_freq": 2,
      "n_grad_time": 4,
      "n_std_thresh": 1.2,
      "prop_decrease": 1.0
    },
      "bg_noise_clip": "kirigami_filters/sample_audio/background.wav"
  },
  "kirigami_filter_config": {
    "lr_threshold": 0.4,
    "lr_phoneme_checkpoint_path": "kirigami_filters/model_checkpoints/scipy_phoneme_filter.ckpt"
  },
  "background_detection_lr_filter_config": {
        "background_lr_threshold": 0.4,
        "bg_lr_checkpoint_path": "kirigami_filters/model_checkpoints/noisy_background_scipy_detector.ckpt"
  }
}