korean_mfa

Model Card for korean_mfa

<!-- Provide a quick summary of what the model is/does. -->

This MFA model is for aligning Korean speech.

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

Model Details

Model Description

<!-- Provide a longer summary of what this model is. -->

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

Uses

<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->

Direct Use

<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->

This model is intended to be used for forced alignment of Korean 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 section addresses misuse, malicious use, and uses that the model will not work well for. -->

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 Korean speech.

Bias, Risks, and Limitations

<!-- This section is meant to convey both technical and sociotechnical 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

<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->

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 korean_mfa dictionary and G2P model

• Source: wikipron
• Orthography: Hangul
• Phone set: MFA
• Words: 79,188

• Phones: 95
• Graphemes: 2,250

IPA chart

Consonants

Vowels

Training Details

Training Data

<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->

Common Voice Korean

• Source: https://voice.mozilla.org/en/datasets
• License: CC-0
• Dialects: N/A
• Number of hours: 1.41
• Number of utterances: 916
• Number of speakers: 46
  • Female speakers: 7
  • Male speakers: 15
  • Unknown speakers: 24

GlobalPhone Korean

• Source: https://catalogue.elra.info/en-us/repository/browse/ELRA-S0200/
• License: ELRA
• Dialects: N/A
• Number of hours: 20.88
• Number of utterances: 8,007
• Number of speakers: 99
  • Female speakers: 50
  • Male speakers: 49
  • Unknown speakers: 0

Deeply Korean read speech corpus public sample

• Source: https://www.openslr.org/97/
• License: CC BY-NC-ND 4.0
• Dialects: N/A
• Number of hours: 2.93
• Number of utterances: 1,697
• Number of speakers: 2
  • Female speakers: 2
  • Male speakers: 0
  • Unknown speakers: 0

ASR-KCSC A Korean Conversational Speech Corpus

• Source: https://magichub.com/datasets/korean-conversational-speech-corpus/
• License: CC BY-NC-ND 4.0
• Dialects: N/A
• Number of hours: 2.38
• Number of utterances: 4,404
• Number of speakers: 14
  • Female speakers: 4
  • Male speakers: 10
  • Unknown speakers: 0

ASR-SKDuSC A Scripted Korean Daily-use Speech Corpus

• Source: https://magichub.com/datasets/korean-scripted-speech-corpus-daily-use-sentence/
• License: CC BY-NC-ND 4.0
• Dialects: N/A
• Number of hours: 3.90
• Number of utterances: 5,240
• Number of speakers: 10
  • Female speakers: 5
  • Male speakers: 5
  • Unknown speakers: 0

Korean Single Speaker Speech Dataset

• Source: https://www.kaggle.com/datasets/bryanpark/korean-single-speaker-speech-dataset
• License: CC BY-NC-SA 4.0
• Dialects: N/A
• Number of hours: 12.86
• Number of utterances: 12,854
• Number of speakers: 40
  • Female speakers: 1
  • Male speakers: 0
  • Unknown speakers: 0

Pansori TEDxKR

• Source: http://www.openslr.org/58/
• License: CC BY-NC-ND 4.0
• Dialects: N/A
• Number of hours: 2.69
• Number of utterances: 2,989
• Number of speakers: 41
  • Female speakers: 9
  • Male speakers: 32
  • Unknown speakers: 0

Zeroth Korean

• Source: https://openslr.org/40/
• License: CC BY 4.0
• Dialects: N/A
• Number of hours: 52.87
• Number of utterances: 22,720
• Number of speakers: 115
  • Female speakers: 70
  • Male speakers: 45
  • Unknown speakers: 0

Training Procedure

<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->

Preprocessing

Preprocessing include fixes and orthographic standardization to various corpora.

Training Hyperparameters

• Training regime: Training configuration

Evaluation

<!-- This section describes the evaluation protocols and provides the results. -->

Testing Data, Factors & Metrics

Testing Data

<!-- This should link to a Dataset Card if possible. -->

N/A

Factors

<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->

N/A

Metrics

<!-- These are the evaluation metrics being used, ideally with a description of why. -->

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

<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->

BibTeX:

@techreport{mfa_korean_mfa_acoustic_2026,
	author={McAuliffe, Michael and Sonderegger, Morgan},
	title={Korean MFA acoustic model v3.3.0},
	address={\url{https://huggingface.co/MontrealCorpusTools/korean_mfa}},
	year={2026},
	month={Jun},
}

APA:

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

Contact

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