preprocessor_config.json

{
  "audio_processor_type": "SeamlessM4TFeatureExtractor",
  "auto_map": {
    "AutoProcessor": "processing_qwen2_mm.Qwen2MMProcessor"
  },
  "do_convert_rgb": true,
  "do_normalize": true,
  "do_rescale": true,
  "do_resize": true,
  "feature_size": 80,
  "image_mean": [
    0.48145466,
    0.4578275,
    0.40821073
  ],
  "image_processor_type": "Qwen2VLImageProcessor",
  "image_std": [
    0.26862954,
    0.26130258,
    0.27577711
  ],
  "max_pixels": 12845056,
  "merge_size": 2,
  "min_pixels": 3136,
  "num_mel_bins": 80,
  "padding_side": "right",
  "padding_value": 0.0,
  "patch_size": 14,
  "processor_class": "Qwen2MMProcessor",
  "resample": 3,
  "rescale_factor": 0.00392156862745098,
  "return_attention_mask": true,
  "sampling_rate": 16000,
  "size": {
    "max_pixels": 12845056,
    "min_pixels": 3136
  },
  "stride": 2,
  "temporal_patch_size": 2
}