paddlespeech.t2s.models.fastspeech2.fastspeech2 module

Fastspeech2 related modules for paddle

class paddlespeech.t2s.models.fastspeech2.fastspeech2.FastSpeech2(idim: int, odim: int, adim: int = 384, aheads: int = 4, elayers: int = 6, eunits: int = 1536, dlayers: int = 6, dunits: int = 1536, postnet_layers: int = 5, postnet_chans: int = 512, postnet_filts: int = 5, postnet_dropout_rate: float = 0.5, positionwise_layer_type: str = 'conv1d', positionwise_conv_kernel_size: int = 1, use_scaled_pos_enc: bool = True, use_batch_norm: bool = True, encoder_normalize_before: bool = True, decoder_normalize_before: bool = True, encoder_concat_after: bool = False, decoder_concat_after: bool = False, reduction_factor: int = 1, encoder_type: str = 'transformer', decoder_type: str = 'transformer', transformer_enc_dropout_rate: float = 0.1, transformer_enc_positional_dropout_rate: float = 0.1, transformer_enc_attn_dropout_rate: float = 0.1, transformer_dec_dropout_rate: float = 0.1, transformer_dec_positional_dropout_rate: float = 0.1, transformer_dec_attn_dropout_rate: float = 0.1, transformer_activation_type: str = 'relu', conformer_pos_enc_layer_type: str = 'rel_pos', conformer_self_attn_layer_type: str = 'rel_selfattn', conformer_activation_type: str = 'swish', use_macaron_style_in_conformer: bool = True, use_cnn_in_conformer: bool = True, zero_triu: bool = False, conformer_enc_kernel_size: int = 7, conformer_dec_kernel_size: int = 31, cnn_dec_dropout_rate: float = 0.2, cnn_postnet_dropout_rate: float = 0.2, cnn_postnet_resblock_kernel_sizes: List[int] = [256, 256], cnn_postnet_kernel_size: int = 5, cnn_decoder_embedding_dim: int = 256, duration_predictor_layers: int = 2, duration_predictor_chans: int = 384, duration_predictor_kernel_size: int = 3, duration_predictor_dropout_rate: float = 0.1, energy_predictor_layers: int = 2, energy_predictor_chans: int = 384, energy_predictor_kernel_size: int = 3, energy_predictor_dropout: float = 0.5, energy_embed_kernel_size: int = 9, energy_embed_dropout: float = 0.5, stop_gradient_from_energy_predictor: bool = False, pitch_predictor_layers: int = 2, pitch_predictor_chans: int = 384, pitch_predictor_kernel_size: int = 3, pitch_predictor_dropout: float = 0.5, pitch_embed_kernel_size: int = 9, pitch_embed_dropout: float = 0.5, stop_gradient_from_pitch_predictor: bool = False, spk_num: Optional[int] = None, spk_embed_dim: Optional[int] = None, spk_embed_integration_type: str = 'add', tone_num: Optional[int] = None, tone_embed_dim: Optional[int] = None, tone_embed_integration_type: str = 'add', init_type: str = 'xavier_uniform', init_enc_alpha: float = 1.0, init_dec_alpha: float = 1.0, enable_speaker_classifier: bool = False, hidden_sc_dim: int = 256)

Bases: Layer

FastSpeech2 module.

This is a module of FastSpeech2 described in FastSpeech 2: Fast and High-Quality End-to-End Text to Speech. Instead of quantized pitch and energy, we use token-averaged value introduced in FastPitch: Parallel Text-to-speech with Pitch Prediction.

Args:

Returns:

Methods

__call__(*inputs, **kwargs)	Call self as a function.
add_parameter(name, parameter)	Adds a Parameter instance.
add_sublayer(name, sublayer)	Adds a sub Layer instance.
apply(fn)	Applies fn recursively to every sublayer (as returned by .sublayers()) as well as self.
buffers([include_sublayers])	Returns a list of all buffers from current layer and its sub-layers.
children()	Returns an iterator over immediate children layers.
clear_gradients()	Clear the gradients of all parameters for this layer.
create_parameter(shape[, attr, dtype, ...])	Create parameters for this layer.
create_tensor([name, persistable, dtype])	Create Tensor for this layer.
create_variable([name, persistable, dtype])	Create Tensor for this layer.
eval()	Sets this Layer and all its sublayers to evaluation mode.
extra_repr()	Extra representation of this layer, you can have custom implementation of your own layer.
forward(text, text_lengths, speech, ...[, ...])	Calculate forward propagation.
full_name()	Full name for this layer, composed by name_scope + "/" + MyLayer.__class__.__name__
inference(text[, durations, pitch, energy, ...])	Generate the sequence of features given the sequences of characters.
load_dict(state_dict[, use_structured_name])	Set parameters and persistable buffers from state_dict.
named_buffers([prefix, include_sublayers])	Returns an iterator over all buffers in the Layer, yielding tuple of name and Tensor.
named_children()	Returns an iterator over immediate children layers, yielding both the name of the layer as well as the layer itself.
named_parameters([prefix, include_sublayers])	Returns an iterator over all parameters in the Layer, yielding tuple of name and parameter.
named_sublayers([prefix, include_self, ...])	Returns an iterator over all sublayers in the Layer, yielding tuple of name and sublayer.
parameters([include_sublayers])	Returns a list of all Parameters from current layer and its sub-layers.
register_buffer(name, tensor[, persistable])	Registers a tensor as buffer into the layer.
register_forward_post_hook(hook)	Register a forward post-hook for Layer.
register_forward_pre_hook(hook)	Register a forward pre-hook for Layer.
set_dict(state_dict[, use_structured_name])	Set parameters and persistable buffers from state_dict.
set_state_dict(state_dict[, use_structured_name])	Set parameters and persistable buffers from state_dict.
state_dict([destination, include_sublayers, ...])	Get all parameters and persistable buffers of current layer and its sub-layers.
sublayers([include_self])	Returns a list of sub layers.
to([device, dtype, blocking])	Cast the parameters and buffers of Layer by the give device, dtype and blocking.
to_static_state_dict([destination, ...])	Get all parameters and buffers of current layer and its sub-layers.
train()	Sets this Layer and all its sublayers to training mode.

backward
encoder_infer
register_state_dict_hook

encoder_infer(text: Tensor, spk_id=None, alpha: float = 1.0, spk_emb=None, tone_id=None) → Tuple[Tensor, Tensor, Tensor]

forward(text: Tensor, text_lengths: Tensor, speech: Tensor, speech_lengths: Tensor, durations: Tensor, pitch: Tensor, energy: Tensor, tone_id: Optional[Tensor] = None, spk_emb: Optional[Tensor] = None, spk_id: Optional[Tensor] = None) → Tuple[Tensor, Dict[str, Tensor], Tensor]

Calculate forward propagation.

Args:

text(Tensor(int64)):

Batch of padded token ids (B, Tmax).

text_lengths(Tensor(int64)):

Batch of lengths of each input (B,).

speech(Tensor):

Batch of padded target features (B, Lmax, odim).

speech_lengths(Tensor(int64)):

Batch of the lengths of each target (B,).

durations(Tensor(int64)):

Batch of padded durations (B, Tmax).

pitch(Tensor):

Batch of padded token-averaged pitch (B, Tmax, 1).

energy(Tensor):

Batch of padded token-averaged energy (B, Tmax, 1).

tone_id(Tensor, optional(int64)):

Batch of padded tone ids (B, Tmax).

spk_emb(Tensor, optional):

Batch of speaker embeddings (B, spk_embed_dim).

spk_id(Tnesor, optional(int64)):

Batch of speaker ids (B,)

Returns:

inference(text: Tensor, durations: Optional[Tensor] = None, pitch: Optional[Tensor] = None, energy: Optional[Tensor] = None, alpha: float = 1.0, use_teacher_forcing: bool = False, spk_emb=None, spk_id=None, tone_id=None) → Tuple[Tensor, Tensor, Tensor]

Generate the sequence of features given the sequences of characters.

Args:

text(Tensor(int64)):

Input sequence of characters (T,).

durations(Tensor, optional (int64)):

Groundtruth of duration (T,).

pitch(Tensor, optional):

Groundtruth of token-averaged pitch (T, 1).

energy(Tensor, optional):

Groundtruth of token-averaged energy (T, 1).

alpha(float, optional):

Alpha to control the speed.

use_teacher_forcing(bool, optional):

Whether to use teacher forcing. If true, groundtruth of duration, pitch and energy will be used.

spk_emb(Tensor, optional, optional):

peaker embedding vector (spk_embed_dim,). (Default value = None)

spk_id(Tensor, optional(int64), optional):

spk ids (1,). (Default value = None)

tone_id(Tensor, optional(int64), optional):

tone ids (T,). (Default value = None)

Returns:

class paddlespeech.t2s.models.fastspeech2.fastspeech2.FastSpeech2Inference(normalizer, model)

Bases: Layer

Methods

__call__(*inputs, **kwargs)	Call self as a function.
add_parameter(name, parameter)	Adds a Parameter instance.
add_sublayer(name, sublayer)	Adds a sub Layer instance.
apply(fn)	Applies fn recursively to every sublayer (as returned by .sublayers()) as well as self.
buffers([include_sublayers])	Returns a list of all buffers from current layer and its sub-layers.
children()	Returns an iterator over immediate children layers.
clear_gradients()	Clear the gradients of all parameters for this layer.
create_parameter(shape[, attr, dtype, ...])	Create parameters for this layer.
create_tensor([name, persistable, dtype])	Create Tensor for this layer.
create_variable([name, persistable, dtype])	Create Tensor for this layer.
eval()	Sets this Layer and all its sublayers to evaluation mode.
extra_repr()	Extra representation of this layer, you can have custom implementation of your own layer.
forward(text[, spk_id, spk_emb])	Defines the computation performed at every call.
full_name()	Full name for this layer, composed by name_scope + "/" + MyLayer.__class__.__name__
load_dict(state_dict[, use_structured_name])	Set parameters and persistable buffers from state_dict.
named_buffers([prefix, include_sublayers])	Returns an iterator over all buffers in the Layer, yielding tuple of name and Tensor.
named_children()	Returns an iterator over immediate children layers, yielding both the name of the layer as well as the layer itself.
named_parameters([prefix, include_sublayers])	Returns an iterator over all parameters in the Layer, yielding tuple of name and parameter.
named_sublayers([prefix, include_self, ...])	Returns an iterator over all sublayers in the Layer, yielding tuple of name and sublayer.
parameters([include_sublayers])	Returns a list of all Parameters from current layer and its sub-layers.
register_buffer(name, tensor[, persistable])	Registers a tensor as buffer into the layer.
register_forward_post_hook(hook)	Register a forward post-hook for Layer.
register_forward_pre_hook(hook)	Register a forward pre-hook for Layer.
set_dict(state_dict[, use_structured_name])	Set parameters and persistable buffers from state_dict.
set_state_dict(state_dict[, use_structured_name])	Set parameters and persistable buffers from state_dict.
state_dict([destination, include_sublayers, ...])	Get all parameters and persistable buffers of current layer and its sub-layers.
sublayers([include_self])	Returns a list of sub layers.
to([device, dtype, blocking])	Cast the parameters and buffers of Layer by the give device, dtype and blocking.
to_static_state_dict([destination, ...])	Get all parameters and buffers of current layer and its sub-layers.
train()	Sets this Layer and all its sublayers to training mode.

backward
register_state_dict_hook

forward(text, spk_id=None, spk_emb=None)

Defines the computation performed at every call. Should be overridden by all subclasses.

Parameters:

*inputs(tuple): unpacked tuple arguments **kwargs(dict): unpacked dict arguments

class paddlespeech.t2s.models.fastspeech2.fastspeech2.FastSpeech2Loss(use_masking: bool = True, use_weighted_masking: bool = False)

Bases: Layer

Loss function module for FastSpeech2.

Methods

__call__(*inputs, **kwargs)	Call self as a function.
add_parameter(name, parameter)	Adds a Parameter instance.
add_sublayer(name, sublayer)	Adds a sub Layer instance.
apply(fn)	Applies fn recursively to every sublayer (as returned by .sublayers()) as well as self.
buffers([include_sublayers])	Returns a list of all buffers from current layer and its sub-layers.
children()	Returns an iterator over immediate children layers.
clear_gradients()	Clear the gradients of all parameters for this layer.
create_parameter(shape[, attr, dtype, ...])	Create parameters for this layer.
create_tensor([name, persistable, dtype])	Create Tensor for this layer.
create_variable([name, persistable, dtype])	Create Tensor for this layer.
eval()	Sets this Layer and all its sublayers to evaluation mode.
extra_repr()	Extra representation of this layer, you can have custom implementation of your own layer.
forward(after_outs, before_outs, d_outs, ...)	Calculate forward propagation.
full_name()	Full name for this layer, composed by name_scope + "/" + MyLayer.__class__.__name__
load_dict(state_dict[, use_structured_name])	Set parameters and persistable buffers from state_dict.
named_buffers([prefix, include_sublayers])	Returns an iterator over all buffers in the Layer, yielding tuple of name and Tensor.
named_children()	Returns an iterator over immediate children layers, yielding both the name of the layer as well as the layer itself.
named_parameters([prefix, include_sublayers])	Returns an iterator over all parameters in the Layer, yielding tuple of name and parameter.
named_sublayers([prefix, include_self, ...])	Returns an iterator over all sublayers in the Layer, yielding tuple of name and sublayer.
parameters([include_sublayers])	Returns a list of all Parameters from current layer and its sub-layers.
register_buffer(name, tensor[, persistable])	Registers a tensor as buffer into the layer.
register_forward_post_hook(hook)	Register a forward post-hook for Layer.
register_forward_pre_hook(hook)	Register a forward pre-hook for Layer.
set_dict(state_dict[, use_structured_name])	Set parameters and persistable buffers from state_dict.
set_state_dict(state_dict[, use_structured_name])	Set parameters and persistable buffers from state_dict.
state_dict([destination, include_sublayers, ...])	Get all parameters and persistable buffers of current layer and its sub-layers.
sublayers([include_self])	Returns a list of sub layers.
to([device, dtype, blocking])	Cast the parameters and buffers of Layer by the give device, dtype and blocking.
to_static_state_dict([destination, ...])	Get all parameters and buffers of current layer and its sub-layers.
train()	Sets this Layer and all its sublayers to training mode.

backward
register_state_dict_hook

forward(after_outs: Tensor, before_outs: Tensor, d_outs: Tensor, p_outs: Tensor, e_outs: Tensor, ys: Tensor, ds: Tensor, ps: Tensor, es: Tensor, ilens: Tensor, olens: Tensor, spk_logits: Optional[Tensor] = None, spk_ids: Optional[Tensor] = None) → Tuple[Tensor, Tensor, Tensor, Tensor, Tensor]

Calculate forward propagation.

Args:

after_outs(Tensor):

Batch of outputs after postnets (B, Lmax, odim).

before_outs(Tensor):

Batch of outputs before postnets (B, Lmax, odim).

d_outs(Tensor):

Batch of outputs of duration predictor (B, Tmax).

p_outs(Tensor):

Batch of outputs of pitch predictor (B, Tmax, 1).

e_outs(Tensor):

Batch of outputs of energy predictor (B, Tmax, 1).

ys(Tensor):

Batch of target features (B, Lmax, odim).

ds(Tensor):

Batch of durations (B, Tmax).

ps(Tensor):

Batch of target token-averaged pitch (B, Tmax, 1).

es(Tensor):

Batch of target token-averaged energy (B, Tmax, 1).

ilens(Tensor):

Batch of the lengths of each input (B,).

olens(Tensor):

Batch of the lengths of each target (B,).

spk_logits(Option[Tensor]):

Batch of outputs after speaker classifier (B, Lmax, num_spk)

spk_ids(Option[Tensor]):

Batch of target spk_id (B,)

Returns:

class paddlespeech.t2s.models.fastspeech2.fastspeech2.StyleFastSpeech2Inference(normalizer, model, pitch_stats_path=None, energy_stats_path=None)

Bases: FastSpeech2Inference

Methods

__call__(*inputs, **kwargs)	Call self as a function.
add_parameter(name, parameter)	Adds a Parameter instance.
add_sublayer(name, sublayer)	Adds a sub Layer instance.
apply(fn)	Applies fn recursively to every sublayer (as returned by .sublayers()) as well as self.
buffers([include_sublayers])	Returns a list of all buffers from current layer and its sub-layers.
children()	Returns an iterator over immediate children layers.
clear_gradients()	Clear the gradients of all parameters for this layer.
create_parameter(shape[, attr, dtype, ...])	Create parameters for this layer.
create_tensor([name, persistable, dtype])	Create Tensor for this layer.
create_variable([name, persistable, dtype])	Create Tensor for this layer.
eval()	Sets this Layer and all its sublayers to evaluation mode.
extra_repr()	Extra representation of this layer, you can have custom implementation of your own layer.
forward(text[, durations, durations_scale, ...])	Args:
full_name()	Full name for this layer, composed by name_scope + "/" + MyLayer.__class__.__name__
load_dict(state_dict[, use_structured_name])	Set parameters and persistable buffers from state_dict.
named_buffers([prefix, include_sublayers])	Returns an iterator over all buffers in the Layer, yielding tuple of name and Tensor.
named_children()	Returns an iterator over immediate children layers, yielding both the name of the layer as well as the layer itself.
named_parameters([prefix, include_sublayers])	Returns an iterator over all parameters in the Layer, yielding tuple of name and parameter.
named_sublayers([prefix, include_self, ...])	Returns an iterator over all sublayers in the Layer, yielding tuple of name and sublayer.
parameters([include_sublayers])	Returns a list of all Parameters from current layer and its sub-layers.
register_buffer(name, tensor[, persistable])	Registers a tensor as buffer into the layer.
register_forward_post_hook(hook)	Register a forward post-hook for Layer.
register_forward_pre_hook(hook)	Register a forward pre-hook for Layer.
set_dict(state_dict[, use_structured_name])	Set parameters and persistable buffers from state_dict.
set_state_dict(state_dict[, use_structured_name])	Set parameters and persistable buffers from state_dict.
state_dict([destination, include_sublayers, ...])	Get all parameters and persistable buffers of current layer and its sub-layers.
sublayers([include_self])	Returns a list of sub layers.
to([device, dtype, blocking])	Cast the parameters and buffers of Layer by the give device, dtype and blocking.
to_static_state_dict([destination, ...])	Get all parameters and buffers of current layer and its sub-layers.
train()	Sets this Layer and all its sublayers to training mode.

backward
denorm
norm
register_state_dict_hook

denorm(data, mean, std)

forward(text: Tensor, durations: Optional[Union[Tensor, ndarray]] = None, durations_scale: Optional[Union[int, float]] = None, durations_bias: Optional[Union[int, float]] = None, pitch: Optional[Union[Tensor, ndarray]] = None, pitch_scale: Optional[Union[int, float]] = None, pitch_bias: Optional[Union[int, float]] = None, energy: Optional[Union[Tensor, ndarray]] = None, energy_scale: Optional[Union[int, float]] = None, energy_bias: Optional[Union[int, float]] = None, robot: bool = False, spk_emb=None, spk_id=None)

Args:

text(Tensor(int64)):

Input sequence of characters (T,).

durations(paddle.Tensor/np.ndarray, optional (int64)):

Groundtruth of duration (T,), this will overwrite the set of durations_scale and durations_bias

durations_scale(int/float, optional):

durations_bias(int/float, optional):

pitch(paddle.Tensor/np.ndarray, optional):

Groundtruth of token-averaged pitch (T, 1), this will overwrite the set of pitch_scale and pitch_bias

pitch_scale(int/float, optional):

In denormed HZ domain.

pitch_bias(int/float, optional):

In denormed HZ domain.

energy(paddle.Tensor/np.ndarray, optional):

Groundtruth of token-averaged energy (T, 1), this will overwrite the set of energy_scale and energy_bias

energy_scale(int/float, optional):

In denormed domain.

energy_bias(int/float, optional):

In denormed domain.

robot(bool) (Default value = False):

spk_emb(Default value = None):

spk_id(Default value = None):

Returns:

Tensor: logmel

norm(data, mean, std)