config.json

{
  "_name_or_path": "av_hubert_hf",
  "activation_dropout": 0.0,
  "activation_function": "relu",
  "adapter_attn_dim": null,
  "adapter_kernel_size": 3,
  "adapter_stride": 2,
  "add_adapter": false,
  "apply_spec_augment": true,
  "architectures": [
    "AV2TextForConditionalGeneration"
  ],
  "attention_dropout": 0.1,
  "audio_dropout": 0.5,
  "audio_feat_dim": 104,
  "bos_token_id": 1,
  "classifier_proj_size": 256,
  "codevector_dim": 256,
  "contrastive_logits_temperature": 0.1,
  "conv_bias": false,
  "conv_channels": 1024,
  "conv_dim": [
    512,
    512,
    512,
    512,
    512,
    512,
    512
  ],
  "conv_kernel": [
    10,
    3,
    3,
    3,
    3,
    2,
    2
  ],
  "conv_kernel_sizes": [
    5,
    5
  ],
  "conv_stride": [
    5,
    2,
    2,
    2,
    2,
    2,
    2
  ],
  "ctc_loss_reduction": "sum",
  "ctc_zero_infinity": false,
  "d_model": 1024,
  "decoder_attention_heads": 8,
  "decoder_ffn_dim": 4096,
  "decoder_layerdrop": 0.0,
  "decoder_layers": 9,
  "decoder_start_token_id": 2,
  "diversity_loss_weight": 0.1,
  "do_stable_layer_norm": false,
  "dropout": 0.1,
  "dropout_features": 0.1,
  "dropout_input": 0.1,
  "encoder_attention_heads": 16,
  "encoder_embed_dim": 1024,
  "encoder_ffn_dim": 2048,
  "encoder_layerdrop": 0.0,
  "encoder_layers": 12,
  "eos_token_id": 2,
  "feat_extract_activation": "gelu",
  "feat_extract_norm": "group",
  "feat_proj_dropout": 0.1,
  "feat_quantizer_dropout": 0.0,
  "feature_grad_mult": 0.1,
  "final_dim": 256,
  "final_dropout": 0.0,
  "freeze_feat_extract_train": true,
  "hidden_act": "gelu",
  "hidden_dropout": 0.1,
  "init_std": 0.02,
  "initializer_range": 0.02,
  "input_channels": 1,
  "input_feat_per_channel": 80,
  "intermediate_size": 4096,
  "is_encoder_decoder": true,
  "label_rate": 25,
  "layer_norm_eps": 1e-05,
  "layerdrop": 0.0,
  "logit_temp": 0.1,
  "mask_channel_length": 10,
  "mask_channel_min_space": 1,
  "mask_channel_other": 0.0,
  "mask_channel_prob": 0.0,
  "mask_channel_selection": "static",
  "mask_feature_length": 10,
  "mask_feature_min_masks": 0,
  "mask_feature_prob": 0.0,
  "mask_length_audio": 10,
  "mask_length_image": 5,
  "mask_min_space": 1,
  "mask_other": 0.0,
  "mask_prob_audio": 0.8,
  "mask_prob_image": 0.3,
  "mask_selection": "static",
  "mask_time_length": 10,
  "mask_time_min_masks": 2,
  "mask_time_min_space": 1,
  "mask_time_other": 0.0,
  "mask_time_prob": 0.0,
  "mask_time_selection": "static",
  "masking_type": "input",
  "max_source_positions": 6000,
  "max_target_positions": 2048,
  "modality_dropout": 0.5,
  "modality_fuse": "concat",
  "model_type": "speech_to_text",
  "no_mask_channel_overlap": false,
  "no_mask_overlap": false,
  "no_mask_time_overlap": false,
  "num_adapter_layers": 3,
  "num_classes": 2004,
  "num_codevector_groups": 2,
  "num_codevectors_per_group": 320,
  "num_conv_layers": 2,
  "num_conv_pos_embedding_groups": 16,
  "num_conv_pos_embeddings": 128,
  "num_dictionaries": 1,
  "num_feat_extract_layers": 7,
  "num_hidden_layers": 24,
  "num_negatives": 100,
  "output_hidden_size": 1024,
  "pad_token_id": 1,
  "proj_codevector_dim": 256,
  "resnet_relu_type": "prelu",
  "resnet_weights": null,
  "sample_rate": 25,
  "scale_embedding": null,
  "selection_type": "same_seq",
  "sim_type": "cosine",
  "skip_masked": false,
  "skip_nomask": false,
  "sub_encoder_layers": 0,
  "target_glu": false,
  "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": "4.44.2",
  "untie_final_proj": true,
  "use_cache": true,
  "use_weighted_layer_sum": false,
  "vocab_size": 1000,
  "xvector_output_dim": 512
}