All commands

The mfa command line utility has several subcommands, which are listed below grouped by general domain.

Preparation

Command	Description	Link
mfa validate	Validate a corpus	Validating data

Forced Alignment

Command	Description	Link
mfa align	Perform forced alignment with a pretrained model	Align with an acoustic model (mfa align)
mfa train	Train an acoustic model and export resulting alignment	Train a new acoustic model (mfa train)
mfa adapt	Adapt a pretrained acoustic model on a new dataset	Adapt acoustic model to new data (mfa adapt)
mfa train_dictionary	Estimate pronunciation probabilities from aligning a corpus	Add probabilities to a dictionary (mfa train_dictionary)

Corpus creation

Command	Description	Link
mfa create_segments	Use voice activity detection to create segments	Segment transcribed files (mfa segment)
mfa train_ivector	Train an ivector extractor for speaker classification	Train an ivector extractor (mfa train_ivector)
mfa diarize_speakers	Use ivector extractor to classify files or cluster them	Speaker diarization (mfa diarize_speakers)
mfa transcribe	Generate transcriptions using an acoustic model, dictionary, and language model	Transcribe audio files (mfa transcribe)
mfa train_lm	Train a language model from a text corpus or from an existing language model	Train a new language model (mfa train_lm)
mfa anchor	Run the Anchor annotator utility (if installed) for editing and managing corpora	Anchor annotator (mfa anchor)

Other utilities

Command	Description	Link
mfa model	Inspect/list/download/save models	Pretrained models
mfa configure	Configure MFA to use customized defaults for command line arguments	Configuration
mfa history	List previous MFA commands run locally

Grapheme-to-phoneme

Command	Description	Link
mfa g2p	Use a G2P model to generate a pronunciation dictionary	Generate pronunciations for words (mfa g2p)
mfa train_g2p	Train a G2P model from a pronunciation dictionary	Train a new G2P model (mfa train_g2p)