Encodes a sentence to token IDs Args: sentence (str): Sentence in text form Returns:

Builds an index for a single audio file's utterances.

Builds an index for a single audio file's utterances.

Builds an index for a single audio file's utterances.

Construct a chunk index for sampling.

Compute the accuracy between two sequences by finding optimal matching. Args: sequence1: A list of integers or strings. sequence2: A list o

Concatenate training data. Args: train_sequences: a list of 2-dim numpy arrays to be concatenated train_cluster_ids: a list of 1-

Decode a list of IDs Args: tokens (list): List of token IDs Returns: str: Output strings

Decode a list of IDs Args: tokens (list): List of token IDs Returns: str: Ou

Estimate the transition bias. Args: cluster_id: Either a list of cluster indicator sequences, or a single concatenated sequen

r""" Actual wandb object. To use wandb features do the following. Example:: self.logger.experiment.some_wandb_fun

Initialize the weights.

r"""Inputs of forward function Args: x: the sequence fed to the positional encoder model (required). Shape:

Args: Audio: Tensor of [batch, audio_len]

Args: x: Melspec features [batch, features, time]

Args: x: Melspec features [batch, features, time]

Check if this list of utterances contain valid audio timestamps.

Produce a string to summarize the experiment.

Pack sequences for training. Args: sub_sequences: A list of numpy array, with obsevation vector from the same cluster in the

Parse arguments. Returns: A tuple of: - `model_args`: model arguments - `training_args`: training arguments - `infer

Compute regularization loss. Args: params: iterable of all parameters weight: weight for the regularization term Returns: the regula

Resize sequences for packing and batching. Args: sequence: (real numpy matrix, size: seq_len*obs_size) - observed sequence cluste

Compute sigma2 prior loss. Args: num_non_zero: since rnn_truth is a collection of different length sequences padded with zeros to fit t

Performs a single optimization step. Arguments: closure (callable, optional): A closure that reevaluates the model

A function to convert string to bool value.

Called at the end of test to aggregate outputs :param outputs: list of individual outputs of each test step :return:

Updates config with data