DeepSpeech¶ class torchaudio.models.DeepSpeech(n_feature: int, n_hidden: int = 2048, n_class: int = 40, dropout: float = 0.0)[source]¶ DeepSpeech 架构,出自论文《Deep Speech: Scaling up end-to-end speech recognition》[Hannun 等人,2014]。 参数: n_feature – 输入特征数量 n_hidden – 内部隐藏单元大小。 n_class – 输出类别数量 方法¶ forward¶ DeepSpeech.forward(x: Tensor) → Tensor[source]¶ 参数: x (torch.Tensor) – 维度为 (batch, channel, time, feature) 的张量。 返回: 维度为 (batch, time, class) 的预测张量。 返回类型: Tensor