Dictionary

class montreal_forced_aligner.db.Dictionary(**kwargs)

Bases: Base

Database class for storing information about a pronunciation dictionary

Parameters:

• id (int) – Primary key

• name (str) – Dictionary name

• dialect (str) – Dialect of dictionary if dictionary name is in MFA format

• path (Path) – Path to the dictionary

• phone_set_type (PhoneSetType) – Phone set

• bracket_regex (str) – Regular expression for detecting bracketed words

• laughter_regex (str) – Regular expression for detecting laughter words

• position_dependent_phones (bool) – Flag for whether phones have word-position flags

• default (bool) – Flag for whether this dictionary is the default one

• clitic_marker (str) – Character marking clitics

• silence_word (str) – Symbol for silence

• optional_silence_phone (str) – Symbol for silence phone

• oov_word (str) – Symbol for unknown words

• bracketed_word (str) – Symbol for bracketed words (cutoffs, hesitations, etc)

• laughter_word (str) – Symbol for laughter words

• max_disambiguation_symbol (int) – Highest disambiguation index required

• silence_probability (float) – Probability of inserting non-initial optional silence

• initial_silence_probability (float) – Probability of inserting initial silence

• final_silence_correction (float) – Correction factor on having final silence

• final_non_silence_correction (float) – Correction factor on having final non-silence

property align_lexicon_disambig_path

Path of lexicon file to use for aligning lattices

property align_lexicon_int_path

Path of lexicon file to use for aligning lattices

property align_lexicon_path

Path of lexicon file to use for aligning lattices

property clitic_set

Set of clitic words

property data_source_identifier

Dictionary name

property disambiguation_symbols_int_path

Path to the word boundary integer IDs

property grapheme_symbol_table_path

Path to file containing grapheme symbols and their integer IDs

property identifier

Dictionary name

property lexicon_disambig_fst_path

Path of disambiguated lexicon fst (L.fst)

property lexicon_fst_path

Path of disambiguated lexicon fst (L.fst)

property phone_disambig_path

Path to file containing phone symbols and their integer IDs

property phone_symbol_table_path

Path to file containing phone symbols and their integer IDs

property phones_directory

Phones directory

property silence_probability_info

Dictionary of silence information

property temp_directory

Path of disambiguated lexicon fst (L.fst)

property word_boundary_int_path

Path to the word boundary integer IDs

property words_symbol_path

Path of word to int mapping file for the dictionary