PretrainedAligner

class montreal_forced_aligner.alignment.PretrainedAligner(acoustic_model_path=None, use_reference_alignments=False, fine_tune=False, fine_tune_boundary_tolerance=None, **kwargs)

Bases: TranscriberMixin, TopLevelMfaWorker

Class for aligning a dataset using a pretrained acoustic model

Parameters:

acoustic_model_path (str) – Path to acoustic model

See also

CorpusAligner

For dictionary and corpus parsing parameters and alignment parameters

TopLevelMfaWorker

For top-level parameters

align()

Run the aligner

align_one_utterance(utterance, session)

Align a single utterance

Parameters:

• utterance (Utterance) – Utterance object to align

• session (Session) – Session to use

property configuration

Configuration for aligner

fine_tune_alignments()

Fine tune aligned boundaries to millisecond precision

fine_tune_arguments()

Generate Job arguments for FineTuneFunction

Returns:

Arguments for processing

Return type:

list[FineTuneArguments]

classmethod parse_parameters(config_path=None, args=None, unknown_args=None)

Parse parameters from a config path or command-line arguments

Parameters:

• config_path (Path) – Config path

• args (dict[str, Any]) – Parsed arguments

• unknown_args (list[str], optional) – Extra command-line arguments

Returns:

Configuration parameters

Return type:

dict[str, Any]

setup()

Setup for alignment

setup_acoustic_model()

Set up the acoustic model