U
    CŒÑhÎ  ã                   @   s,   d dl mZmZ dgZG dd„ dejƒZdS )é    )ÚnnÚTensorÚ
Wav2Letterc                       s>   e Zd ZdZdeeeddœ‡ fdd„Zeed	œd
d„Z‡  Z	S )r   au  Wav2Letter model architecture from *Wav2Letter: an End-to-End ConvNet-based Speech
    Recognition System* :cite:`collobert2016wav2letter`.

    See Also:
        * `Training example <https://github.com/pytorch/audio/tree/release/0.12/examples/pipeline_wav2letter>`__

    Args:
        num_classes (int, optional): Number of classes to be classified. (Default: ``40``)
        input_type (str, optional): Wav2Letter can use as input: ``waveform``, ``power_spectrum``
         or ``mfcc`` (Default: ``waveform``).
        num_features (int, optional): Number of input features that the network will receive (Default: ``1``).
    é(   Úwaveformé   N)Únum_classesÚ
input_typeÚnum_featuresÚreturnc                    s¦  t ƒ  ¡  |dkrdn|}t tj|ddddd�tjdd�tjddd	d
dd�tjdd�tjddd	d
dd�tjdd�tjddd	d
dd�tjdd�tjddd	d
dd�tjdd�tjddd	d
dd�tjdd�tjddd	d
dd�tjdd�tjddd	d
dd�tjdd�tjdddd
dd�tjdd�tjddd
d
dd�tjdd�tjd|d
d
dd�tjdd�¡}|dk�r’t tj|ddddd�tjdd�¡}t ||¡| _|dk�r¢|| _d S )Nr   éú   é0   é   é   )Zin_channelsZout_channelsZkernel_sizeZstrideÚpaddingT)Zinplaceé   r   é   iÐ  é    é   r   é    é-   )Zpower_spectrumZmfcc)ÚsuperÚ__init__r   Z
SequentialZConv1dZReLUÚacoustic_model)Úselfr   r	   r
   Zacoustic_num_featuresr   Zwaveform_model©Ú	__class__© úJ/tmp/user/2202/pip-unpacked-wheel-p89lpdx5/torchaudio/models/wav2letter.pyr      sD    











ê

þ
zWav2Letter.__init__)Úxr   c                 C   s   |   |¡}tjj|dd�}|S )zæ
        Args:
            x (torch.Tensor): Tensor of dimension (batch_size, num_features, input_length).

        Returns:
            Tensor: Predictor tensor of dimension (batch_size, number_of_classes, input_length).
        r   )Zdim)r   r   Z
functionalZlog_softmax)r   r   r   r   r   Úforward=   s    	
zWav2Letter.forward)r   r   r   )
Ú__name__Ú
__module__Ú__qualname__Ú__doc__ÚintÚstrr   r   r    Ú__classcell__r   r   r   r   r      s   'N)Ztorchr   r   Ú__all__ÚModuler   r   r   r   r   Ú<module>   s   ÿ