@inproceedings{simmons-2025-data,
    title = "Data augmentation for low-resource bilingual {ASR} from {T}ira linguistic elicitation using Whisper",
    author = "Simmons, Mark",
    editor = "Lachler, Jordan  and
      Agyapong, Godfred  and
      Arppe, Antti  and
      Moeller, Sarah  and
      Chaudhary, Aditi  and
      Rijhwani, Shruti  and
      Rosenblum, Daisy",
    booktitle = "Proceedings of the Eight Workshop on the Use of Computational Methods in the Study of Endangered Languages",
    month = mar,
    year = "2025",
    address = "Honolulu, Hawaii, USA",
    publisher = "Association for Computational Linguistics",
    url = "https://preview.aclanthology.org/Ingest-2025-COMPUTEL/2025.computel-main.18/",
    pages = "155--161",
    ISBN = "None",
    abstract = "This paper explores finetuning Whisper for transcribing audio from linguistic elicitation of Tira, a Heiban language of Sudan. Audio originates from linguistic fieldwork and is bilingual in English and Tira. We finetune Whisper large-v3 using hand-labeled Tira audio and evaluate the resulting model on bilingual audio. We show that Whisper exhibits catastrophic forgetting of English after only a small amount of training, but that including automatically annotated English spans of audio in the training data dramatically reduces catastrophic forgetting of English while largely preserving ASR performance on monolingual Tira audio. This work is relevant to the study of automatic speech recognition for under-resourced languages and for contexts of bilingualism in a high and low-resourced language."
}