Abstract: The recognition of accented speech still remains a dominant problem in Automatic Speech Recognition (ASR) systems. We approach the classification of accented English speech through the Emphasized Channel Attention, Propagation and Aggregation Time Delay Neural Network (ECAPA-TDNN) architecture which has been shown to perform well on a variety of speech tasks. Three models are proposed: one trained from scratch, another two models (one using data augmentation and a baseline model) fine-tuned from the checkpoints of speechbrain/spkrec-ecapa-voxceleb (VoxCeleb). Our results show that the model fine-tuned with data augmentation yield the best results. Most of the misclassifications were structured and expected due to accent similarities, such as the American and Canadian accents. We also explored the internal categorization of embeddings through t-SNE, a dimensionality reduction technique, and found that there was a level of clustering based on phonological similarity. For future work, we would like to explore the implementation of this accent classification system in our suggested framework to improve ASR performance by making it more inclusive to accented speech.

This repository provides all the necessary tools to perform accent identification from speech recordings with SpeechBrain. The system uses a model pretrained on the CommonAccent dataset in English (16 accents). This system is based on the CommonLanguage Recipe located here: https://github.com/speechbrain/speechbrain/tree/develop/recipes/CommonLanguage

The provided system can recognize the following 16 accents from short speech recordings in English (EN):

african
australia
bermuda
canada
england
hongkong
indian
ireland
malaysia
newzealand
philippines
scotland
singapore
southatlandtic
us
wales

Github repository link: https://github.com/JuanPZuluaga/accent-recog-slt2022

For a better experience, we encourage you to learn more about SpeechBrain. The given model performance on the test set is:

This system is composed of an ECAPA model coupled with statistical pooling. A classifier, trained with Categorical Cross-Entropy Loss, is applied on top of that.

The system is trained with recordings sampled at 16kHz (single channel). The code will automatically normalize your audio (i.e., resampling + mono channel selection) when calling classify_file if needed. Make sure your input tensor is compliant with the expected sampling rate if you use encode_batch and classify_batch.

First of all, please install SpeechBrain with the following command:

Please notice that we encourage you to read our tutorials and learn more about SpeechBrain.

To perform inference on the GPU, add run_opts={"device":"cuda"} when calling the from_hparams method.

The model was trained with SpeechBrain.

To train it from scratch follow these steps:

You can find our training results (models, logs, etc) in this repository's Files and versions page.

The SpeechBrain team does not provide any warranty on the performance achieved by this model when used on other datasets.

If you find useful this work, please cite our work as:

@article{zuluaga2023commonaccent,
  title={CommonAccent: Exploring Large Acoustic Pretrained Models for Accent Classification Based on Common Voice},
  author={Zuluaga-Gomez, Juan and Ahmed, Sara and Visockas, Danielius and Subakan, Cem},
  journal={Interspeech 2023},
  url={https://arxiv.org/abs/2305.18283},
  year={2023}
}

Please, cite SpeechBrain if you use it for your research or business.