r"""Load CLAP model with args from config file

Load a CLIP model, preserve its text pretrained part, and set in the CLAP model Parameters ---------- name : str A model name lis

Make mask tensor containing indices of padded part. Args: lengths (LongTensor or List): Batch of lengths (B,). xs (Tensor, optiona

Replace non-padding symbols with their position numbers. Position numbers begin at padding_idx+1. Padding symbols are ignored.

median_filter :param x: input prediction array of shape (B, T, C) or (B, T). Input is a sequence of probabilities 0 <= x <= 1 :param w

:return: {'item_name': Str, 'wav_fn': Str, 'txt': Str, 'spk_name': Str, 'txt_loader': None or Func}

decayed_noise = noise_decay(beta, f0mean) decayed_noise = n[t]exp(-t * f_mean / beta / samp_rate) beta: (dim=1) or (batchsize=1, 1,

Use the PLMS method from [Pseudo Numerical Methods for Diffusion Models on Manifolds](https://arxiv.org/abs/2202.09778).

Pack waveform and target of several audio clips to a single hdf5 file. This can speed up loading and training.

Pad all audio to specific length.