--- language: - en pipeline_tag: automatic-speech-recognition library_name: transformers tags: - arxiv:2605.03297 - wav2vec2 - ctc - supervised-contrastive-learning - accented-speech - l2-arctic license: other --- # Wav2Vec2 Large L2-ARCTIC SupCon Repeated: 8-Fold Split 0 Associated paper: [Contrastive Regularization for Accent-Robust ASR](https://arxiv.org/abs/2605.03297). English accented-speech recognition model trained with CTC and supervised contrastive learning. Repeated L2-ARCTIC prompts provide positive SupCon pairs. This checkpoint was trained using split 0 of the repository's 8-fold setup. The included processor supports greedy CTC decoding. ## Data and Code - L2-ARCTIC: https://psi.engr.tamu.edu/l2-arctic-corpus/ - Training code and split metadata: https://github.com/thaivanphat95/robust-atc-asr ## Usage ```python import torch import soundfile as sf from transformers import AutoModelForCTC, AutoProcessor model_id = "thaivanphat95/wav2vec2-large-l2-arctic-supcon-repeated-8fold-0" processor = AutoProcessor.from_pretrained(model_id) model = AutoModelForCTC.from_pretrained(model_id).eval() audio, sample_rate = sf.read("audio.wav") inputs = processor(audio, sampling_rate=sample_rate, return_tensors="pt") with torch.no_grad(): logits = model(**inputs).logits print(processor.batch_decode(torch.argmax(logits, dim=-1))[0]) ``` Audio should be mono and sampled at 16 kHz. ## Citation ```bibtex @article{thai2026contrastive, title={Contrastive Regularization for Accent-Robust ASR}, author={Thai, Van-Phat and Dhruv, Aradhya and Pham, Duc-Thinh and Alam, Sameer}, journal={arXiv preprint arXiv:2605.03297}, year={2026}, doi={10.48550/arXiv.2605.03297} } ``` ## Limitations and License The model is specialized for L2-ARCTIC-style English read speech and should not be used without review in safety-critical settings. Model-weight use and redistribution may be affected by the licenses of the pretrained model and L2-ARCTIC. The training code's Apache-2.0 license does not cover these weights.
