convert_ali_to_textgrids

aligner.multiprocessing.convert_ali_to_textgrids(output_directory, model_directory, dictionary, corpus, num_jobs)[source]

Multiprocessing function that aligns based on the current model

See:

for more details on the Kaldi binaries this function calls.

Also see https://github.com/kaldi-asr/kaldi/blob/master/egs/wsj/s5/steps/get_train_ctm.sh for the bash script that this function was based on.

Parameters:
output_directorystr

Directory to write TextGrid files to

model_directorystr

Directory of training (monophone, triphone, speaker-adapted triphone training directories)

dictionaryDictionary

Dictionary object that has information about pronunciations

corpusCorpus

Corpus object that has information about the dataset

num_jobsint

The number of processes to use in calculation

Raises:
CorpusError

If the files per speaker exceeds the number of files that are allowed to be open on the computer (for Unix-based systems)