Synthesizer for Training

Residual Vector Quantizer. Args: dimension (int): Dimension of the codebooks. n_q (int): Number of residual vector quantizers used

BigVGAN is a neural vocoder model that applies anti-aliased periodic activation for residual blocks (resblocks). New in BigVGAN-v2: it can op

Synthesizer for Training

Zero-pads model inputs and targets

1) loads audio, speaker_id, text pairs 2) normalizes text and converts them to sequences of integers 3) computes spectrograms from audio

Implements Warmup learning rate schedule until 'warmup_steps', going from 'init_lr' to 'peak_lr' for multiple optimizers.

r"""Adaptive Layer Normalization

r"""Adaptive Layer Normalization

Conv1d + GLU(Gated Linear Unit) with residual connection. For GLU refer to https://arxiv.org/abs/1612.08083 paper.

ConvNeXt Block adapted from https://github.com/facebookresearch/ConvNeXt to 1D audio signal. Args: dim (int): Number of input channels.

Maintain similar input lengths in a batch. Length groups are specified by boundaries. Ex) boundaries = [b1, b2, b3] -> any batch is inclu

Implements 'Scaled Adam', a variant of Adam where we scale each parameter's update proportional to the norm of that parameter; and also lea

Implementation of a sine-based periodic activation function Shape: - Input: (B, C, T) - Output: (B, C, T), same shape as the

A modified Snake function which uses separate parameters for the magnitude of the periodic components Shape: - Input: (B, C, T)

Synthesizer for Training

dataset class for text tokens to semantic model training.

Modifies the backpropped derivatives of a function to try to encourage, for each channel, that it is positive at least a proportion `threshol

Dialted and Depth-Separable Convolution