This model is intended to be used for forced alignment of English varieties that it was trained on (US, UK, Indian, Nigerian Englishes). Please see https://montreal-forced-aligner.readthedocs.io/en/latest/user_guide/troubleshooting.html for details on common fixes.

Out-of-Scope Use

This model cannot provide accurate assessments of goodness of pronunciations or provide transcripts as it is trained to be accepting of variation in pronunciation to provide a reasonable alignment across many varieties of English.

Bias, Risks, and Limitations

This model will perform best on the variety of speech that it was trained on, with a bias towards US English. The speakers in the training data are all adult speakers, so child speech alignment may not be accurate.

Recommendations

When using this model on a variety that it was not trained on, better results can be attained by adapting the model to the data to be aligned first. See https://montreal-forced-aligner.readthedocs.io/en/latest/user_guide/workflows/adapt_acoustic_model.html and https://github.com/mmcauliffe/mfa-adaptation for example usage and scripts.

How to Get Started with the Model

Use the code below to get started with the model.

To get started, follow the instructions for installing MFA. To align files using this model, use the mfa align command.

Dictionary Details

Details for english_nigeria_mfa dictionary and G2P model

Source: wikipron
Orthography: Latin
Phone set: MFA
Words: 77,407

Phones: 66
Graphemes: 46

IPA chart

Consonants

Manner	Labial	Labiodental	Dental	Alveolar	Alveopalatal	Palatal	Velar	Glottal
Nasal	m	ɱ		n		ɲ	ŋ
Palatalized	mʲ
Stop	p b		t̪ d̪	t d		c ɟ	k ɡ
Aspirated	pʰ			tʰ		cʰ	kʰ
Labialized				tʷ		cʷ ɟʷ	kʷ ɡʷ
Palatalized	pʲ bʲ			tʲ dʲ
Affricate					tʃ dʒ
Sibilant				s z	ʃ
Fricative		f v	θ ð			ç		h
Palatalized		fʲ vʲ
Approximant	w			ɹ		j
Lateral				l ɫ		ʎ

Vowels

	Front	Central	Near-Back	Back
Close	i iː			u uː
			ʊ
Close-Mid	e			o

Open-Mid	ɛ ɛː	ɜ ɜː		ɔ

Open		a aː

Diphthongs

Other

Details for english_india_mfa dictionary and G2P model

Source: wikipron
Orthography: Latin
Phone set: MFA
Words: 123,926

Phones: 64
Graphemes: 62

IPA chart

Consonants

Manner	Labial	Labiodental	Dental	Alveolar	Alveopalatal	Retroflex	Palatal	Velar	Glottal
Nasal	m	ɱ		n			ɲ	ŋ
Palatalized	mʲ
Stop	p b		t̪ d̪			ʈ ɖ	c ɟ	k ɡ	ʔ
Labialized	pʷ					ʈʷ	cʷ ɟʷ	kʷ ɡʷ
Palatalized	pʲ bʲ					ʈʲ
Affricate					tʃ dʒ
Sibilant				s z	ʃ ʒ
Fricative		f					ç		h
Palatalized		fʲ
Approximant		ʋ		ɹ			j
Tap				ɾ
Lateral				l			ʎ

Vowels

	Front	Near-Front	Central	Near-Back	Back
Close	i iː		ʉ ʉː
		ɪ		ʊ
Close-Mid	eː				oː
			ə
Open-Mid	ɛ ɛː		ɜ ɜː

Open			a aː		ɑ ɑː ɒ ɒː

Diphthongs

Details for english_uk_mfa dictionary and G2P model

Source: wikipron
Orthography: Latin
Phone set: MFA
Words: 186,963

Phones: 78
Graphemes: 88

IPA chart

Consonants

Manner	Labial	Labiodental	Dental	Alveolar	Alveopalatal	Palatal	Velar	Glottal
Nasal	m m̩	ɱ		n n̩		ɲ	ŋ
Palatalized	mʲ
Stop	p b		t̪ d̪	t d		c ɟ	k ɡ	ʔ
Aspirated	pʰ			tʰ		cʰ	kʰ
Labialized	pʷ			tʷ		cʷ ɟʷ	kʷ ɡʷ
Palatalized	pʲ bʲ			tʲ dʲ
Affricate					tʃ dʒ
Sibilant				s z	ʃ ʒ
Fricative		f v	θ ð			ç		h
Labialized		vʷ
Palatalized		fʲ vʲ
Approximant	w			ɹ		j
Lateral				l ɫ ɫ̩		ʎ

Vowels

	Front	Near-Front	Central	Near-Back	Back
Close	i iː		ʉ ʉː
		ɪ		ʊ
Close-Mid	e ej
			ə
Open-Mid	ɛ ɛː		ɜ ɜː
			ɐ
Open			a		ɑ ɑː ɒ ɒː

Diphthongs

Details for english_us_mfa dictionary and G2P model

Source: wikipron
Orthography: Latin
Phone set: MFA
Words: 251,336

Phones: 78
Graphemes: 87

IPA chart

Consonants

Manner	Labial	Labiodental	Dental	Alveolar	Alveopalatal	Palatal	Velar	Glottal
Nasal	m m̩	ɱ		n n̩		ɲ	ŋ
Palatalized	mʲ
Stop	p b		t̪ d̪	t d		c ɟ	k ɡ	ʔ
Aspirated	pʰ			tʰ		cʰ	kʰ
Labialized	pʷ			tʷ		cʷ ɟʷ	kʷ ɡʷ
Palatalized	pʲ bʲ			tʲ dʲ
Affricate					tʃ dʒ
Sibilant				s z	ʃ ʒ
Fricative		f v	θ ð			ç		h
Palatalized		fʲ vʲ
Approximant	w			ɹ		j
Tap				ɾ ɾ̃
Palatalized				ɾʲ
Lateral				l ɫ ɫ̩		ʎ

Vowels

	Front	Near-Front	Central	Near-Back	Back
Close	i iː		ʉ ʉː
		ɪ		ʊ
Close-Mid	ej				ow
			ə ɚ
Open-Mid	ɛ		ɝ
	æ		ɐ
Open					ɑ ɑː ɒ ɒː

Diphthongs

Training Details

Training Data

ARU English corpus

Source: http://datacat.liverpool.ac.uk/681/
License: CC BY 3.0
Dialects: uk
Number of hours: 7.13
Number of utterances: 8,640
Number of speakers: 12
- Female speakers: 0
- Male speakers: 0
- Unknown speakers: 0

Common Voice English

Source: https://voice.mozilla.org/en/datasets
License: CC-0
Dialects: us, uk, nigeria, india
Number of hours: 2,322.80
Number of utterances: 1,625,987
Number of speakers: 71,160
- Female speakers: 3,750
- Male speakers: 14,586
- Unknown speakers: 52,824

Corpus of Regional African American Language

Source: https://oraal.uoregon.edu/coraal
License: CC BY-NC-SA 4.0
Dialects: us
Number of hours: 124.31
Number of utterances: 236,792
Number of speakers: 193
- Female speakers: 88
- Male speakers: 94
- Unknown speakers: 11

Google Nigerian English

Source: https://openslr.org/70/
License: CC BY-SA 4.0
Dialects: nigeria
Number of hours: 5.77
Number of utterances: 3,359
Number of speakers: 31
- Female speakers: 19
- Male speakers: 12
- Unknown speakers: 0

Google UK and Ireland English

Source: https://openslr.org/83/
License: CC BY-SA 4.0
Dialects: uk
Number of hours: 31.29
Number of utterances: 17,877
Number of speakers: 120
- Female speakers: 49
- Male speakers: 71
- Unknown speakers: 0

ICE-Nigeria

Source: https://sourceforge.net/projects/ice-nigeria/
License: CC BY-NC-SA 3.0
Dialects: nigeria
Number of hours: 52.86
Number of utterances: 113,664
Number of speakers: 1,276
- Female speakers: 293
- Male speakers: 853
- Unknown speakers: 130

L2-ARCTIC

Source: https://psi.engr.tamu.edu/l2-arctic-corpus/
License: CC BY-NC 4.0
Dialects: N/A
Number of hours: 27.51
Number of utterances: 27,042
Number of speakers: 24
- Female speakers: 12
- Male speakers: 12
- Unknown speakers: 0

LibriSpeech English

Source: https://openslr.org/12/
License: CC BY 4.0
Dialects: us
Number of hours: 982.10
Number of utterances: 292,367
Number of speakers: 2,484
- Female speakers: 1,283
- Male speakers: 1,201
- Unknown speakers: 0

NCHLT English

Source: https://repo.sadilar.org/handle/20.500.12185/274
License: CC BY 3.0
Dialects: N/A
Number of hours: 56.43
Number of utterances: 77,412
Number of speakers: 210
- Female speakers: 0
- Male speakers: 0
- Unknown speakers: 210

A Scripted Pakistani English Daily-use Speech Corpus

Source: https://magichub.com/datasets/pakistani-english-scripted-speech-corpus-daily-use-sentence/
License: CC BY-NC-ND 4.0
Dialects: india
Number of hours: 4.00
Number of utterances: 2,191
Number of speakers: 7
- Female speakers: 3
- Male speakers: 4
- Unknown speakers: 0

Training Procedure

Preprocessing

Preprocessing include fixes and orthographic standardization to various corpora.

Training Hyperparameters

Training regime: Training configuration

Evaluation

Testing Data, Factors & Metrics

Testing Data

N/A

Factors

N/A

Metrics

N/A

Results

N/A

Summary

Technical Specifications

Model Architecture and Objective

HMM-GMM model

Software

This model was trained via the Montreal Forced Aligner.

Citation

BibTeX:

@techreport{mfa_english_mfa_acoustic_2026,
	author={McAuliffe, Michael and Sonderegger, Morgan},
	title={Global English MFA acoustic model v3.3.0},
	address={\url{https://huggingface.co/MontrealCorpusTools/english_mfa}},
	year={2026},
	month={Jun},
}

APA:

McAuliffe, M. & Sonderegger, M. (2026). Global English MFA acoustic model v3.3.0. Available at https://huggingface.co/MontrealCorpusTools/english_mfa.

Contact

For questions and issues, please file an issue either for this model at https://huggingface.co/MontrealCorpusTools/english_mfa/discussions or for larger MFA issues at https://github.com/MontrealCorpusTools/Montreal-Forced-Aligner/issues.