kaira.losses.audio.MultiResolutionSTFTLoss

Inheritance diagram of MultiResolutionSTFTLoss — Inheritance diagram for MultiResolutionSTFTLoss

class kaira.losses.audio.MultiResolutionSTFTLoss(fft_sizes=[512, 1024, 2048], hop_sizes=[128, 256, 512], win_lengths=[512, 1024, 2048], window='hann')[source]

Bases: BaseLoss

Multi-Resolution STFT Loss Module.

This module calculates STFT loss at multiple resolutions for better time-frequency coverage.

Methods

`__init__`	Initialize the MultiResolutionSTFTLoss module.
`forward`	Forward pass through the MultiResolutionSTFTLoss module.

__init__(fft_sizes=[512, 1024, 2048], hop_sizes=[128, 256, 512], win_lengths=[512, 1024, 2048], window='hann')[source]

Initialize the MultiResolutionSTFTLoss module.

Parameters:

fft_sizes (list) – List of FFT sizes for each resolution. Default is [512, 1024, 2048].
hop_sizes (list) – List of hop sizes for each resolution. Default is [128, 256, 512].
win_lengths (list) – List of window lengths for each resolution. Default is [512, 1024, 2048].
window (str) – Window function type. Default is ‘hann’.

forward(x: Tensor, target: Tensor) → Tensor[source]

Forward pass through the MultiResolutionSTFTLoss module.

Parameters:

x (torch.Tensor) – The input audio tensor.
target (torch.Tensor) – The target audio tensor.

Returns:

The multi-resolution STFT loss.

Return type:

torch.Tensor