config.json

{
  "architectures": [
    "FastSpeech2ConformerModel"
  ],
  "decoder_config": {
    "attention_dropout_rate": 0.2,
    "concat_after": false,
    "dropout_rate": 0.2,
    "kernel_size": 31,
    "layers": 4,
    "linear_units": 1536,
    "normalize_before": false,
    "num_attention_heads": 2,
    "positional_dropout_rate": 0.2
  },
  "duration_predictor_channels": 256,
  "duration_predictor_dropout_rate": 0.2,
  "duration_predictor_kernel_size": 3,
  "duration_predictor_layers": 2,
  "encoder_config": {
    "attention_dropout_rate": 0.2,
    "concat_after": false,
    "dropout_rate": 0.2,
    "kernel_size": 7,
    "layers": 4,
    "linear_units": 1536,
    "normalize_before": false,
    "num_attention_heads": 2,
    "positional_dropout_rate": 0.2
  },
  "encoder_layers": 4,
  "encoder_num_attention_heads": 2,
  "energy_embed_dropout": 0.0,
  "energy_embed_kernel_size": 1,
  "energy_predictor_channels": 256,
  "energy_predictor_dropout": 0.5,
  "energy_predictor_kernel_size": 3,
  "energy_predictor_layers": 2,
  "hidden_size": 384,
  "input_dim": 78,
  "is_encoder_decoder": true,
  "max_source_positions": 5000,
  "model_type": "fastspeech2_conformer",
  "num_languages": null,
  "num_mel_bins": 80,
  "num_speakers": null,
  "pitch_embed_dropout": 0.0,
  "pitch_embed_kernel_size": 1,
  "pitch_predictor_channels": 256,
  "pitch_predictor_dropout": 0.5,
  "pitch_predictor_kernel_size": 5,
  "pitch_predictor_layers": 5,
  "positionwise_conv_kernel_size": 3,
  "reduction_factor": 1,
  "speaker_embed_dim": null,
  "speaking_speed": 1.0,
  "speech_decoder_postnet_dropout": 0.5,
  "speech_decoder_postnet_kernel": 5,
  "speech_decoder_postnet_layers": 5,
  "speech_decoder_postnet_units": 256,
  "stop_gradient_from_energy_predictor": false,
  "stop_gradient_from_pitch_predictor": true,
  "torch_dtype": "float32",
  "transformers_version": "4.33.0.dev0",
  "use_cnn_in_conformer": true,
  "use_macaron_style_in_conformer": true,
  "use_masking": true,
  "use_weighted_masking": false,
  "vocab_size": 78
}