DictionaryTextCorpusMixin

class montreal_forced_aligner.corpus.text_corpus.DictionaryTextCorpusMixin(**kwargs)

Bases: TextCorpusMixin, MultispeakerDictionaryMixin

Abstract mixin class for processing text corpora with pronunciation dictionaries.

This is primarily useful for training language models, as you can treat words in the language model as OOV if they aren’t in your pronunciation dictionary

See also

TextCorpusMixin

For corpus parsing parameters

MultispeakerDictionaryMixin

For dictionary parsing parameters

load_corpus()

Load the corpus