torchaudio.models.hubert_pretrain_large¶
- torchaudio.models.hubert_pretrain_large(encoder_projection_dropout: float = 0.0, encoder_attention_dropout: float = 0.0, encoder_ff_interm_dropout: float = 0.0, encoder_dropout: float = 0.0, encoder_layer_drop: float = 0.0, mask_prob: float = 0.8, mask_channel_prob: float = 0.0, mask_channel_length: int = 10, feature_grad_mult: Optional[float] = None) HuBERTPretrainModel[source]¶
构建“大型”
HuBERTPretrainModel来自 *HuBERT* [Hsu 等人，2021] 用于预训练。- 参数：
encoder_projection_dropout (float) – 参见
hubert_pretrain_model().encoder_attention_dropout (float) – 参见
hubert_pretrain_model().encoder_ff_interm_dropout (float) – 参见
hubert_pretrain_model().encoder_dropout (float) – 参见
hubert_pretrain_model().encoder_layer_drop (float) – 参见
hubert_pretrain_model().mask_prob (float) – 参见
hubert_pretrain_model().mask_channel_prob (float) – 参见
hubert_pretrain_model().mask_channel_length (int) – 参见
hubert_pretrain_model().feature_grad_mult (float or None) – 参见
hubert_pretrain_model().
- 返回：
生成的模型。
- 返回类型：
