ttslearn.dnntts.gen.predict_acoustic

ttslearn.dnntts.gen.predict_acoustic(device, labels, acoustic_model, acoustic_config, acoustic_in_scaler, acoustic_out_scaler, binary_dict, numeric_dict, mlpg=True)[source]

Predict acoustic features.

Parameters
  • device (torch.device) – pytorch device

  • labels (list) – list of labels

  • acoustic_model (nn.Module) – trained acoustic model

  • acoustic_config (dict) – configuration of acoustic model

  • acoustic_in_scaler (sklearn.preprocessing.StandardScaler) – StandardScaler of acoustic features

  • acoustic_out_scaler (sklearn.preprocessing.StandardScaler) – StandardScaler of acoustic output

  • binary_dict (dict) – dictionary of binary features

  • numeric_dict (dict) – dictionary of numeric features

  • mlpg (bool) – whether to use MLPG

Returns

predicted acoustic features

Return type

numpy.ndarray