config.json

{
  "architectures": [
    "Effb2TrmCaptioningModel"
  ],
  "attn_emb_dim": 1408,
  "decoder_dropout": 0.2,
  "decoder_emb_dim": 256,
  "decoder_n_layers": 2,
  "decoder_we_tie_weights": true,
  "fc_emb_dim": 1408,
  "sample_rate": 16000,
  "shared_dim": 1024,
  "tchr_dim": 768,
  "torch_dtype": "float32",
  "transformers_version": "4.30.2",
  "vocab_size": 4981
}