---
language:
- en
license: apache-2.0
base_model: openai/whisper-small
tags:
- generated_from_trainer
datasets:
- Shiry/ATC_combined
metrics:
- wer
model-index:
- name: Whisper Small ATC - ATCText
  results:
  - task:
      name: Automatic Speech Recognition
      type: automatic-speech-recognition
    dataset:
      name: ATC
      type: Shiry/ATC_combined
      args: 'split: test'
    metrics:
    - name: Wer
      type: wer
      value: 10.612930650580948
---

<!-- This model card has been generated automatically according to the information the Trainer had access to. You
should probably proofread and complete it, then remove this comment. -->

# Whisper Small ATC - ATCText

This model is a fine-tuned version of [openai/whisper-small](https://huggingface.co/openai/whisper-small) on the ATC dataset.
It achieves the following results on the evaluation set:
- Loss: 0.2486
- Wer: 10.6129

## Model description

More information needed

## Intended uses & limitations

More information needed

## Training and evaluation data

More information needed

## Training procedure

### Training hyperparameters

The following hyperparameters were used during training:
- learning_rate: 1e-05
- train_batch_size: 8
- eval_batch_size: 8
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_steps: 500
- training_steps: 4000

### Training results

| Training Loss | Epoch | Step | Validation Loss | Wer     |
|:-------------:|:-----:|:----:|:---------------:|:-------:|
| 0.2533        | 0.42  | 1000 | 0.3465          | 16.2868 |
| 0.235         | 0.84  | 2000 | 0.2881          | 13.5237 |
| 0.0851        | 1.27  | 3000 | 0.2607          | 10.6048 |
| 0.1317        | 1.69  | 4000 | 0.2486          | 10.6129 |


### Framework versions

- Transformers 4.39.3
- Pytorch 2.2.2
- Datasets 2.18.0
- Tokenizers 0.15.2


### Additional Information
## Licensing Information
The licensing status of the dataset hinges on the legal status of the UWB-ATCC corpus creators.

They used Creative Commons - Attribution-NonCommercial-ShareAlike 4.0 International (CC BY-NC-SA 4.0) licensing.

## Citation Information
Contributors who prepared, processed, normalized and uploaded the dataset in HuggingFace:

@article{zuluaga2022how,
    title={How Does Pre-trained Wav2Vec2. 0 Perform on Domain Shifted ASR? An Extensive Benchmark on Air Traffic Control Communications},
    author={Zuluaga-Gomez, Juan and Prasad, Amrutha and Nigmatulina, Iuliia and Sarfjoo, Saeed and others},
    journal={IEEE Spoken Language Technology Workshop (SLT), Doha, Qatar},
    year={2022}
  }

@article{zuluaga2022bertraffic,
  title={BERTraffic: BERT-based Joint Speaker Role and Speaker Change Detection for Air Traffic Control Communications},
  author={Zuluaga-Gomez, Juan and Sarfjoo, Seyyed Saeed and Prasad, Amrutha and others},
  journal={IEEE Spoken Language Technology Workshop (SLT), Doha, Qatar},
  year={2022}
  }

@article{zuluaga2022atco2,
  title={ATCO2 corpus: A Large-Scale Dataset for Research on Automatic Speech Recognition and Natural Language Understanding of Air Traffic Control Communications},
  author={Zuluaga-Gomez, Juan and Vesel{\`y}, Karel and Sz{\"o}ke, Igor and Motlicek, Petr and others},
  journal={arXiv preprint arXiv:2211.04054},
  year={2022}
}

## Authors of the dataset:

@article{vsmidl2019air,
  title={Air traffic control communication (ATCC) speech corpora and their use for ASR and TTS development},
  author={{\v{S}}m{\'\i}dl, Lubo{\v{s}} and {\v{S}}vec, Jan and Tihelka, Daniel and Matou{\v{s}}ek, Jind{\v{r}}ich and Romportl, Jan and Ircing, Pavel},
  journal={Language Resources and Evaluation},
  volume={53},
  number={3},
  pages={449--464},
  year={2019},
  publisher={Springer}
}