{
  "_name_or_path": "Bingsu/torchaudio_tacotron2",
  "architectures": [
    "Tacotron2Model"
  ],
  "attention_dropout": 0.1,
  "attention_hidden_dim": 128,
  "attention_location_kernel_size": 31,
  "attention_location_n_filter": 32,
  "attention_rnn_dim": 1024,
  "auto_map": {
    "AutoConfig": "modeling_tacotron2.Tacotron2Config",
    "AutoModel": "modeling_tacotron2.Tacotron2Model",
    "AutoModelForPreTraining": "modeling_tacotron2.Tacotron2ForPreTraining"
  },
  "decoder_dropout": 0.1,
  "decoder_early_stopping": true,
  "decoder_max_step": 2000,
  "decoder_rnn_dim": 1024,
  "encoder_embedding_dim": 512,
  "encoder_kernel_size": 5,
  "encoder_n_convolution": 3,
  "gate_threshold": 0.5,
  "mask_padding": false,
  "n_frames_per_step": 1,
  "n_mels": 80,
  "n_symbol": 392,
  "postnet_embedding_dim": 512,
  "postnet_kernel_size": 5,
  "postnet_n_convolution": 5,
  "prenet_dim": 256,
  "symbol_embedding_dim": 512,
  "torch_dtype": "float32",
  "transformers_version": "4.25.1"
}