paddlespeech.t2s.datasets.am_batch_fn module

class paddlespeech.t2s.datasets.am_batch_fn.ErnieSATCollateFn(mlm_prob: float = 0.8, mean_phn_span: int = 8, seg_emb: bool = False, text_masking: bool = False)

Bases: object

Functor class of common_collate_fn()

Methods

__call__(exmaples)

Call self as a function.

class paddlespeech.t2s.datasets.am_batch_fn.StarGANv2VCCollateFn(latent_dim: int = 16, max_mel_length: int = 192)

Bases: object

Functor class of common_collate_fn()

Methods

__call__(exmaples)

Call self as a function.

random_clip
starganv2_vc_batch_fn

random_clip(mel: array)

starganv2_vc_batch_fn(examples)

paddlespeech.t2s.datasets.am_batch_fn.build_erniesat_collate_fn(mlm_prob: float = 0.8, mean_phn_span: int = 8, seg_emb: bool = False, text_masking: bool = False)

paddlespeech.t2s.datasets.am_batch_fn.build_starganv2_vc_collate_fn(latent_dim: int = 16, max_mel_length: int = 192)

paddlespeech.t2s.datasets.am_batch_fn.diffsinger_multi_spk_batch_fn(examples)

paddlespeech.t2s.datasets.am_batch_fn.diffsinger_single_spk_batch_fn(examples)

paddlespeech.t2s.datasets.am_batch_fn.erniesat_batch_fn(examples, mlm_prob: float = 0.8, mean_phn_span: int = 8, seg_emb: bool = False, text_masking: bool = False)

paddlespeech.t2s.datasets.am_batch_fn.fastspeech2_multi_spk_batch_fn(examples)

paddlespeech.t2s.datasets.am_batch_fn.fastspeech2_multi_spk_batch_fn_static(examples)

paddlespeech.t2s.datasets.am_batch_fn.fastspeech2_single_spk_batch_fn(examples)

paddlespeech.t2s.datasets.am_batch_fn.fastspeech2_single_spk_batch_fn_static(examples)

paddlespeech.t2s.datasets.am_batch_fn.jets_multi_spk_batch_fn(examples)

Returns:

Dict[str, Any]:

• text (Tensor): Text index tensor (B, T_text).

• text_lengths (Tensor): Text length tensor (B,).

• feats (Tensor): Feature tensor (B, T_feats, aux_channels).

• feats_lengths (Tensor): Feature length tensor (B,).

• durations (Tensor): Feature tensor (B, T_text,).

• durations_lengths (Tensor): Durations length tensor (B,).

• pitch (Tensor): Feature tensor (B, pitch_length,).

• energy (Tensor): Feature tensor (B, energy_length,).

• speech (Tensor): Speech waveform tensor (B, T_wav).

• spk_id (Optional[Tensor]): Speaker index tensor (B,) or (B, 1).

• spk_emb (Optional[Tensor]): Speaker embedding tensor (B, spk_embed_dim).

paddlespeech.t2s.datasets.am_batch_fn.jets_single_spk_batch_fn(examples)

Returns:

Dict[str, Any]:

• text (Tensor): Text index tensor (B, T_text).

• text_lengths (Tensor): Text length tensor (B,).

• feats (Tensor): Feature tensor (B, T_feats, aux_channels).

• feats_lengths (Tensor): Feature length tensor (B,).

• durations (Tensor): Feature tensor (B, T_text,).

• durations_lengths (Tensor): Durations length tensor (B,).

• pitch (Tensor): Feature tensor (B, pitch_length,).

• energy (Tensor): Feature tensor (B, energy_length,).

• speech (Tensor): Speech waveform tensor (B, T_wav).

paddlespeech.t2s.datasets.am_batch_fn.speedyspeech_multi_spk_batch_fn(examples)

paddlespeech.t2s.datasets.am_batch_fn.speedyspeech_multi_spk_batch_fn_static(examples)

paddlespeech.t2s.datasets.am_batch_fn.speedyspeech_single_spk_batch_fn(examples)

paddlespeech.t2s.datasets.am_batch_fn.speedyspeech_single_spk_batch_fn_static(examples)

paddlespeech.t2s.datasets.am_batch_fn.tacotron2_multi_spk_batch_fn(examples)

paddlespeech.t2s.datasets.am_batch_fn.tacotron2_single_spk_batch_fn(examples)

paddlespeech.t2s.datasets.am_batch_fn.transformer_single_spk_batch_fn(examples)

paddlespeech.t2s.datasets.am_batch_fn.vits_multi_spk_batch_fn(examples)

Returns:

Dict[str, Any]:

• text (Tensor): Text index tensor (B, T_text).

• text_lengths (Tensor): Text length tensor (B,).

• feats (Tensor): Feature tensor (B, T_feats, aux_channels).

• feats_lengths (Tensor): Feature length tensor (B,).

• speech (Tensor): Speech waveform tensor (B, T_wav).

• spk_id (Optional[Tensor]): Speaker index tensor (B,) or (B, 1).

• spk_emb (Optional[Tensor]): Speaker embedding tensor (B, spk_embed_dim).

paddlespeech.t2s.datasets.am_batch_fn.vits_single_spk_batch_fn(examples)

Returns:

Dict[str, Any]:

• text (Tensor): Text index tensor (B, T_text).

• text_lengths (Tensor): Text length tensor (B,).

• feats (Tensor): Feature tensor (B, T_feats, aux_channels).

• feats_lengths (Tensor): Feature length tensor (B,).

• speech (Tensor): Speech waveform tensor (B, T_wav).