WordType#
- class montreal_forced_aligner.data.WordType(*values)[source]#
Bases:
Enum
Enum for types of words
- bracketed = 5#
Words that are in brackets
- clitic = 2#
Clitics that must attach to words
- cutoff = 6#
Words that are cutoffs of particular words or hesitations of the next word
- disambiguation = 10#
Disambiguation symbols internal to Kaldi
- interjection = 11#
Set of words that can be added on the fly to transcripts
- laughter = 7#
Words that represent laughter
- music = 9#
Words that represent music
- noise = 8#
Words that represent non-speech noise
- oov = 4#
Words representing out of vocabulary items
- silence = 3#
Words representing silence
- speech = 1#
General speech words