torchaudio.models.hubert_pretrain_xlarge¶
- torchaudio.models.hubert_pretrain_xlarge(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]¶
构建用于预训练的 “超大”
HuBERTPretrainModel，该模型基于 HuBERT [Hsu 等, 2021]。- 参数：
encoder_projection_dropout (float) – 请参阅
hubert_pretrain_model()。encoder_attention_dropout (float) – 请参阅
hubert_pretrain_model()。encoder_ff_interm_dropout (float) – 请参阅
hubert_pretrain_model()。encoder_dropout (float) – 请参阅
hubert_pretrain_model()。encoder_layer_drop (float) – 请参阅
hubert_pretrain_model()。mask_prob (float) – 请参阅
hubert_pretrain_model()。mask_channel_prob (float) – 请参阅
hubert_pretrain_model()。mask_channel_length (int) – 请参阅
hubert_pretrain_model()。feature_grad_mult (float 或 None) – 请参阅
hubert_pretrain_model()。
- 返回：
返回生成的模型。
- 返回类型：
