espnet.nets.tts_interface.TTSInterface

source

class espnet.nets.tts_interface.TTSInterface

Bases: object

TTS Interface for ESPnet model implementation.

Initilize TTS module.

static add_arguments(parser)

Add model specific argments to parser.

property attention_plot_class

Plot attention weights.

property base_plot_keys

Return base key names to plot during training.

The keys should match what chainer.reporter reports. if you add the key loss, the reporter will report main/loss and validation/main/loss values. also loss.png will be created as a figure visulizing main/loss and validation/main/loss values.

• Returns: Base keys to plot during training.
• Return type: list[str]

calculate_all_attentions(*args, **kwargs)

Calculate TTS attention weights.

• Parameters:Tensor – Batch of attention weights (B, Lmax, Tmax).

forward(*args, **kwargs)

Calculate TTS forward propagation.

• Returns: Loss value.
• Return type: Tensor

inference(*args, **kwargs)

Generate the sequence of features given the sequences of characters.

• Returns: The sequence of generated features (L, odim). Tensor: The sequence of stop probabilities (L,). Tensor: The sequence of attention weights (L, T).
• Return type: Tensor

load_pretrained_model(model_path)

Load pretrained model parameters.