preprocessor_config.json

{
    "auto_map": {
        "AutoImageProcessor": "image_processing_bailingmm.BailingMMImageProcessor",
        "AutoFeatureExtractor": "audio_processing_bailingmm.BailingMMAudioProcessor",
        "AutoProcessor": "processing_bailingmm.BailingMMProcessor"
    },
    "image_grid_thws": [1, 32, 32],
    "min_pixels": 78400,
    "max_pixels": 2007040,
    "patch_size": 14,
    "temporal_patch_size": 2,
    "merge_size": 2,
    "image_mean": [
        0.48145466,
        0.4578275,
        0.40821073
    ],
    "image_std": [
        0.26862954,
        0.26130258,
        0.27577711
    ],
    "image_token": "<image>",
    "video_token": "<video>",
    "image_processor_type": "BailingMMImageProcessor",
    "audio_processor_type": "BailingMMAudioProcessor",
    "audio_token": "<audio>",
    "chunk_length": 30,
    "dither": 0.0,
    "feature_size": 80,
    "frame_length": 25,
    "frame_shift": 10,
    "inverse_norm": true,
    "lfr_m": 7,
    "lfr_n": 6,
    "n_samples": 480000,
    "num_audio_tokens": 256,
    "sampling_rate": 16000,
    "up_sample": true,
    "return_attention_mask": true,
    "padding_side": "right",
    "padding_value": 0.0,
    "processor_class": "BailingMMProcessor",
    "wav_frontend_args": {
        "cmvn_file": "am.mvn",
        "fs": 16000,
        "window": "hamming",
        "n_mels": 80,
        "frame_length": 25,
        "frame_shift": 10,
        "lfr_m": 7,
        "lfr_n": 6
    },
    "whisper_frontend_args": {
        "n_mels": 128
    }
}