GlotLID

Description

GlotLID is a Fasttext language identification (LID) model that supports more than 2000 labels.

Latest: GlotLID v4 alpha (model_v4_alpha.bin) is out as a pre-release of the next version: 2056 labels, a smaller model (783 MB instead of 1.69 GB), trained with fastText-turbo. It is not final yet; its labels and scores may still change, so model.bin stays v3 for now. See versions.

Stable: GlotLID V3 supports 2102 labels (three-letter ISO codes with script). For more details on the supported languages and performance, as well as significant changes from previous versions, please refer to https://github.com/cisnlp/GlotLID/blob/main/languages-v3.md.

• Demo: huggingface (language identification, and code-switching with MaskLID)
• Repository: github
• Paper: paper (EMNLP 2023)
• Point of Contact: amir@cis.lmu.de

How to use

Here is how to use this model to detect the language of a given text:

# pip install fasttext-turbo huggingface_hub
>>> import fasttext
>>> from huggingface_hub import hf_hub_download

# model.bin is the latest version always
>>> model_path = hf_hub_download(repo_id="cis-lmu/glotlid", filename="model.bin")
>>> model = fasttext.load_model(model_path)
>>> model.predict("Hello, world!")

We recommend fastText-turbo (pip install fasttext-turbo): it is a drop-in replacement for fastText (same import fasttext, same model files and API, prebuilt wheels) that predicts with GlotLID about 3 times faster than the original fastText in our tests (2.7x with v4 alpha, 3.4-4.3x with v3), with the same predictions. The original fasttext package works too.

To use the v4 pre-release, download model_v4_alpha.bin instead of model.bin.

To limit the prediction to a smaller set of languages without training another model, use CustomLID (see the GitHub README).

If you are not a fan of huggingface_hub, then download the model directly:

>>> ! wget https://huggingface.co/cis-lmu/glotlid/resolve/main/model.bin

>>> import fasttext

>>> model = fasttext.load_model("/path/to/model.bin")
>>> model.predict("Hello, world!")

License

The model is distributed under the Apache License, Version 2.0 plus notices (see LICENSE file for full terms).

Version

We always maintain the previous version of GlotLID in our repository.

To access a specific version, simply append the version number to the filename.

• For v1: model_v1.bin (introduced in the GlotLID paper and used in all experiments).
• For v2: model_v2.bin (an edited version of v1, featuring more languages, and cleaned from noisy corpora based on the analysis of v1).
• For v3: model_v3.bin (an edited version of v2, featuring more languages, excluding macro languages, further cleaned from noisy corpora and incorrect metadata labels based on the analysis of v2, supporting "zxx" and "und" series labels)
• For v4 alpha: model_v4_alpha.bin (a pre-release of v4; not final, its labels and scores may still change). It supports 2056 labels (1899 three-letter ISO codes, with the "zxx" and "und" series). Trained with fastText-turbo; it loads in fastText and fastText-turbo like the previous versions. It is smaller (96-dimensional, 783 MB) and uses character n-grams of 2 to 5 with word n-grams of 1, so nothing is shared between neighbouring words: the scores of a text add up from the scores of its words, which makes the per-word scores that MaskLID uses for code-switching exact.

model.bin always refers to the latest stable version (v3).

References

If you use this model, please cite the following paper:

@inproceedings{
  kargaran2023glotlid,
  title={{GlotLID}: Language Identification for Low-Resource Languages},
  author={Kargaran, Amir Hossein and Imani, Ayyoob and Yvon, Fran{\c{c}}ois and Sch{\"u}tze, Hinrich},
  booktitle={The 2023 Conference on Empirical Methods in Natural Language Processing},
  year={2023},
  url={https://openreview.net/forum?id=dl4e3EBz5j}
}