cjli's picture
Update README.md
00a47d7 verified
metadata
license: cc-by-nc-sa-2.0
task_categories:
  - automatic-speech-recognition
tags:
  - phone-recognition
  - l2-speech
pretty_name: GMU Speech Accent Archive
language:
  - en
configs:
  - config_name: default
    data_files:
      - split: test
        path: data/test-*
dataset_info:
  features:
    - name: id
      dtype: string
    - name: audio
      dtype: audio
    - name: transcript
      dtype: string
  splits:
    - name: test
      num_bytes: 3121706755
      num_examples: 1242
  download_size: 3064680707
  dataset_size: 3121706755

Speech Accent Archive

This work is licensed under CC BY-NC-SA 2.0. If you use this dataset, please remember to cite:

@article{gmuspeechaccentarchive,
    title={Speech Accent Archive},
    author={Weinberger, Steven},
    year={2015},
    publisher={George Mason University},
    note={Retrieved from \url{https://accent.gmu.edu}}
}

Speech Accent Archive is a dataset containing 1000+ utterances from speakers of diverse background. Every speaker read the same short paragraph, and then the speech is transcribed phonetically:

Please call Stella. Ask her to bring these things with her from the store: Six spoons of fresh snow peas, five thick slabs of blue cheese, and maybe a snack for her brother Bob. We also need a small plastic snake and a big toy frog for the kids. She can scoop these things into three red bags, and we will go meet her Wednesday at the train station.

The official website provides interactive search of additional speaker information, while we only include paired speech and phonetic transcription here. (Original files could be obtained via contacting accent@gmu.edu.) The utterance ID is kept the same, and transcripts are extracted from the RTF files. Note that the speech is sampled at 44.1kHz!

This dataset is included as pr-saa in our phone recognition benchmarking toolkit, 💎PRiSM.

@misc{prism2026,
      title={PRiSM: Benchmarking Phone Realization in Speech Models}, 
      author={Shikhar Bharadwaj and Chin-Jou Li and Yoonjae Kim and Kwanghee Choi and Eunjung Yeo and Ryan Soh-Eun Shim and Hanyu Zhou and Brendon Boldt and Karen Rosero Jacome and Kalvin Chang and Darsh Agrawal and Keer Xu and Chao-Han Huck Yang and Jian Zhu and Shinji Watanabe and David R. Mortensen},
      year={2026},
      eprint={2601.14046},
      archivePrefix={arXiv},
      primaryClass={cs.CL},
      url={https://arxiv.org/abs/2601.14046}, 
}