TokenizerModel

class montreal_forced_aligner.models.TokenizerModel(source, root_directory=None)

Bases: Archive

Class for Tokenizer models

Parameters:

• source (str) – Path to source archive

• root_directory (str) – Path to save exported model

add_graphemes_path(source_directory)

Add symbols file into archive

Parameters:

source_directory (Path) – Source directory path

add_meta_file(g2p_trainer)

Construct metadata information for the G2P model from the dictionary it was trained from

Parameters:

g2p_trainer (G2PTrainer) – Trainer for the G2P model

add_tokenizer_model(source_directory)

Add FST file into archive

Parameters:

source_directory (Path) – Source directory path

export_fst_model(destination)

Extract FST model path to destination

Parameters:

destination (Path) – Destination directory

property fst_path

Tokenizer model’s FST path

property input_sym_path

Tokenizer model’s input symbols path

property meta

Metadata for the G2P model

property output_sym_path

Tokenizer model’s output symbols path

property sym_path

Tokenizer model’s grapheme symbols path

validate(*args)

Placeholder