German ASR

This model is trained on the Mozilla Common Voice 8.0, the Spoken Wikipedia Corpus and the m-ailabs corpus. In contrast to our first model this one can transcribe German Umlauts. Additionally it was trained on a newer Version of the Mozilla Common Voice.

We do not provide a language model.

Performance

This model has a WER of 7.09% and is slightly better then the model from our paper: https://huggingface.co/jfreiwa/asr-crdnn-german

Model application

Install SpeechBrain

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

pip install speechbrain

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

Using the model

from speechbrain.pretrained import EncoderDecoderASR

asr_model = EncoderDecoderASR.from_hparams(source="jfreiwa/asr-crdnn-german-umlaute", savedir="pretrained_models/asr-crdnn-german-umlaute")
asr_model.transcribe_file("jfreiwa/asr-crdnn-german/example-de.wav")

Inference on GPU

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

Limitations

We do not provide any warranty on the performance achieved by this model when used on other datasets.

About SpeechBrain

Website: https://speechbrain.github.io/
Code: https://github.com/speechbrain/speechbrain/
HuggingFace: https://huggingface.co/speechbrain/

Citing SpeechBrain

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

@misc{speechbrain,
  title={{SpeechBrain}: A General-Purpose Speech Toolkit},
  author={Mirco Ravanelli and Titouan Parcollet and Peter Plantinga and Aku Rouhe and Samuele Cornell and Loren Lugosch and Cem Subakan and Nauman Dawalatabad and Abdelwahab Heba and Jianyuan Zhong and Ju-Chieh Chou and Sung-Lin Yeh and Szu-Wei Fu and Chien-Feng Liao and Elena Rastorgueva and François Grondin and William Aris and Hwidong Na and Yan Gao and Renato De Mori and Yoshua Bengio},
  year={2021},
  eprint={2106.04624},
  archivePrefix={arXiv},
  primaryClass={eess.AS},
  note={arXiv:2106.04624}
}

Citing our paper

Please, cite our original paper, when you use our models in your research.

@inproceedings{freiwald2022,
  author={J. Freiwald and P. Pracht and S. Gergen and D. Kolossa},
  title={Open-Source End-To-End Learning for Privacy-Preserving German {ASR}},
  year=2022,
  booktitle={DAGA 2022}
}

Acknowledgements

This work was funded by the German Federal Ministry of Education and Research (BMBF) within the “Innovations for Tomorrow’s Production, Services, and Work” Program (02L19C200), a project that is implemented by the Project Management Agency Karlsruhe (PTKA). The authors are responsible for the content of this publication.