Model Card for swedish_mfa

This MFA model is for aligning Swedish speech.

• Model details
• Uses
• Performance Factors
• Dictionary Details
• Training Details
• Evaluation
• Contact

Model Details

Model Description

• Developed by: Michael McAuliffe
• Funded by: N/A
• Model type: Montreal Forced Aligner model
• Language(s) (NLP): Swedish
• License: cc-by-4.0

Uses

Direct Use

This model is intended to be used for forced alignment of Swedish speech. 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 for Swedish speech.

Bias, Risks, and Limitations

This model will perform best on the variety of speech that it was trained on. 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 swedish_mfa dictionary and G2P model

• Source: wikipron
• Orthography: Latin
• Phone set: MFA
• Words: 147,701

• Phones: 44
• Graphemes: 47

IPA chart

Consonants

Manner	Labial	Labiodental	Dental	Alveolar	Retroflex	Palatal	Velar	Glottal
Nasal	m		n̪		ɳ		ŋ
Stop	p b		t̪ d̪		ʈ ɖ		k ɡ
Aspirated	pʰ		t̪ʰ		ʈʰ		kʰ
Sibilant			s̪		ʂ	ɕ
Fricative		f						h
Approximant		ʋ				j
Trill				r
Lateral			l̪
Lateral Tap					ɭ

Vowels

	Front	Near-Front	Central	Near-Back	Back
Close	iː yː		ʉː		uː
		ɪ ʏ		ʊ
Close-Mid	eː øː		ɵ		oː
Open-Mid	ɛ ɛː œ				ɔ
Open			a		ɑː

Other

• ɧ

Training Details

Training Data

Common Voice Swedish

• Source: https://voice.mozilla.org/en/datasets
• License: CC-0
• Dialects: N/A
• Number of hours: 39.97
• Number of utterances: 37,302
• Number of speakers: 674
  • Female speakers: 19
  • Male speakers: 128
  • Unknown speakers: 527

GlobalPhone Swedish

• Source: https://catalogue.elra.info/en-us/repository/browse/ELRA-S0204/
• License: ELRA
• Dialects: N/A
• Number of hours: 21.69
• Number of utterances: 11,816
• Number of speakers: 98
  • Female speakers: 48
  • Male speakers: 50
  • Unknown speakers: 0

NST Swedish

• Source: https://www.nb.no/sprakbanken/en/resource-catalogue/oai-nb-no-sbr-56/
• License: CC-0
• Dialects: N/A
• Number of hours: 378.33
• Number of utterances: 276,372
• Number of speakers: 895
  • Female speakers: 482
  • Male speakers: 407
  • Unknown speakers: 6

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_swedish_mfa_acoustic_2026,
    author={McAuliffe, Michael and Sonderegger, Morgan},
    title={Swedish MFA acoustic model v3.3.0},
    address={\url{https://huggingface.co/MontrealCorpusTools/swedish_mfa}},
    year={2026},
    month={Jun},
}

APA:

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

Contact

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