config.json

{
  "activation_dropout": 0.0,
  "adapter_attn_dim": null,
  "adapter_kernel_size": 3,
  "adapter_stride": 2,
  "add_adapter": false,
  "apply_spec_augment": true,
  "architectures": [
    "CustomWav2Vec2ForSequenceClassification"
  ],
  "attention_dropout": 0.0,
  "bos_token_id": 1,
  "classifier_proj_size": 256,
  "codevector_dim": 768,
  "contrastive_logits_temperature": 0.1,
  "conv_bias": true,
  "conv_dim": [
    512,
    512,
    512,
    512,
    512,
    512,
    512
  ],
  "conv_kernel": [
    10,
    3,
    3,
    3,
    3,
    2,
    2
  ],
  "conv_stride": [
    5,
    2,
    2,
    2,
    2,
    2,
    2
  ],
  "ctc_loss_reduction": "mean",
  "ctc_zero_infinity": false,
  "diversity_loss_weight": 0.1,
  "do_stable_layer_norm": true,
  "eos_token_id": 2,
  "feat_extract_activation": "gelu",
  "feat_extract_dropout": 0.0,
  "feat_extract_norm": "layer",
  "feat_proj_dropout": 0.0,
  "feat_quantizer_dropout": 0.0,
  "final_dropout": 0.0,
  "gradient_checkpointing": false,
  "hidden_act": "gelu",
  "hidden_dropout": 0.0,
  "hidden_size": 1024,
  "id2label": {
    "0": "door open",
    "1": "lights off",
    "2": "alarm on",
    "3": "door close",
    "4": "speaker off",
    "5": "alarm off",
    "6": "refrigerator off",
    "7": "speaker decrease",
    "8": "fan off",
    "9": "fan increase",
    "10": "lights on",
    "11": "fan on",
    "12": "refrigerator on",
    "13": "camera off",
    "14": "tv on",
    "15": "fan decrease",
    "16": "tv off",
    "17": "speaker increase",
    "18": "camera on",
    "19": "speaker on"
  },
  "initializer_range": 0.02,
  "intermediate_size": 4096,
  "label2id": {
    "alarm off": 5,
    "alarm on": 2,
    "camera off": 13,
    "camera on": 18,
    "door close": 3,
    "door open": 0,
    "fan decrease": 15,
    "fan increase": 9,
    "fan off": 8,
    "fan on": 11,
    "lights off": 1,
    "lights on": 10,
    "refrigerator off": 6,
    "refrigerator on": 12,
    "speaker decrease": 7,
    "speaker increase": 17,
    "speaker off": 4,
    "speaker on": 19,
    "tv off": 16,
    "tv on": 14
  },
  "layer_norm_eps": 1e-05,
  "layerdrop": 0.0,
  "mask_feature_length": 10,
  "mask_feature_min_masks": 0,
  "mask_feature_prob": 0.0,
  "mask_time_length": 10,
  "mask_time_min_masks": 2,
  "mask_time_prob": 0.05,
  "model_type": "wav2vec2",
  "num_adapter_layers": 3,
  "num_attention_heads": 16,
  "num_codevector_groups": 2,
  "num_codevectors_per_group": 320,
  "num_conv_pos_embedding_groups": 16,
  "num_conv_pos_embeddings": 128,
  "num_feat_extract_layers": 7,
  "num_hidden_layers": 24,
  "num_negatives": 100,
  "output_hidden_size": 1024,
  "pad_token_id": 27,
  "proj_codevector_dim": 768,
  "tdnn_dilation": [
    1,
    2,
    3,
    1,
    1
  ],
  "tdnn_dim": [
    512,
    512,
    512,
    512,
    1500
  ],
  "tdnn_kernel": [
    5,
    3,
    3,
    1,
    1
  ],
  "torch_dtype": "float32",
  "transformers_version": "4.51.3",
  "use_weighted_layer_sum": false,
  "vocab_size": 30,
  "xvector_output_dim": 512
}