@inproceedings{chambers-etal-2026-addressing,
    title = "Addressing Domain Mismatch in {ASR} for Akuzipik Language Documentation",
    author = "Chambers, Summer  and
      Woodrose Schwartz, Sylvia  and
      Kelley, Matthew  and
      Woodrose Schwartz, Lane",
    editor = "Agyapong, Godfred  and
      Moeller, Sarah  and
      Arppe, Antti  and
      Marashian, Ali  and
      Rosenblum, Daisy",
    booktitle = "Proceedings of the Ninth Workshop on the Use of Computational Methods in the Study of Endangered Languages ({C}omput{EL}-9)",
    month = jul,
    year = "2026",
    address = "San Diego, California, USA",
    publisher = "Association for Computational Linguistics",
    url = "https://preview.aclanthology.org/ingest-acl-workshops/2026.computel-1.10/",
    pages = "93--103",
    ISBN = "979-8-89176-422-4",
    abstract = "The use of ASR models in endangered language documentation has grown in popularity given the bottleneck of manual speech transcription. Meta{'}s Massively Multilingual Speech (MMS) model is particularly popular for its extensibility to low-resource languages. However, it is mostly trained on read speech data from the Bible, meaning it may not perform well on other domains. We evaluated this model on data collected as part of a larger language documentation and revitalization project focused on Akuzipik, a polysynthetic Alaska Native language. We also finetuned and evaluated the model on a small (1h) collection of speech. The original model performed well on a dataset that roughly matched the Bible training data in domain and writing style but struggled on a separate collection of spontaneous speech. Performance on spontaneous speech improved after finetuning on a sample of our full dataset, and error rates reduced less dramatically after finetuning only on read speech. Both finetuning scenarios show promise for future model improvement, especially considering the relative ease of collecting read speech data. This experiment confirms the challenge of transcribing spontaneous speech with the MMS ASR model but provides hope for improving model performance for language documentation purposes, even with scarce data."
}