{
  "architectures": [
    "ASTForAudioClassification"
  ],
  "attention_probs_dropout_prob": 0.1,
  "dtype": "float32",
  "frequency_stride": 2,
  "hidden_act": "gelu",
  "hidden_dropout_prob": 0.1,
  "hidden_size": 32,
  "initializer_range": 0.02,
  "intermediate_size": 37,
  "is_decoder": false,
  "layer_norm_eps": 1e-12,
  "max_length": 24,
  "model_type": "audio-spectrogram-transformer",
  "num_attention_heads": 4,
  "num_hidden_layers": 2,
  "num_mel_bins": 16,
  "patch_size": 2,
  "qkv_bias": true,
  "time_stride": 2,
  "transformers_version": "5.15.0.dev0"
}