Fraunhofer Institute for Systems and Innovation Research ISI

Mayumi Ohta

Researcher

Mayumi Ohta obtained her Bachelor's and Master's degree in Computational Linguistics at Heidelberg University. Her Master thesis "Sparse Perturbations for Improved Convergence in SZO Optimization" received a distinction and also a special mention in the best paper award of the 6th International Conference on Machine Learning, Optimization, and Data Science (LOD). After her graduation, she gathered professional experience both in industry and academia in the field of Natural Language Processing.

Since 2022 she has been a PhD candidate at Heidelberg University. Her current research interests lie in Privacy-Preserving Machine Learning with a special focus on the applications to Machine Translation and Speech Recognition. As a research scientist, she is involved in various projects at the Innovation System Data Excellence Center (ISDEC), Fraunhofer Institute in Karlsruhe.

Expand all Close all

Publications (selection)
Ohta, M.; Kreutzer, J.; Riezler, S. (2022): JoeyS2T: Minimalistic Speech-to-Text Modeling with JoeyNMT, Proceedings of the 2022 Conference on Empirical Methods in Natural Language Processing (EMNLP): System Demonstrations, Abu Dhabi, UAE.

Lam, T.K.; Ohta, M.; Schamoni, S.; Riezler, S. (2021): On-the-Fly Aligned Data Augmentation for Sequence-to-Sequence ASR, Proceedings of the 22th Annual Conference of the International Speech Communication Association (INTERSPEECH), Brno, Czech Republic.

Ohta, M.; Berger, N.; Sokolov, A.; Riezler, S. (2020): Sparse Perturbations for Improved Convergence in Stochastic Zeroth-Order Optimization, The 6th International Conference on Machine Learning, Optimization, and Data Science (LOD), Siena, Italy.

Sokolov, A.; Hitschler, J.; Ohta, M.; Riezler, S. (2018): Sparse Stochastic Zeroth-Order Optimization with an Application to Bandit Structured Prediction, arXiv preprint arXiv:1806.04458, 2018.

Hitschler, J.; Jehl, L.; Karimova, S.; Ohta, M.; Körner, M.; Riezler, S. (2016): Otedama: Fast Rule-based Pre-Ordering for Machine Translation, The Prague Bulletin of Mathematical Linguistics (PBML), 106, 159–168.
- more publications can be found in the database Fraunhofer-PublicaSearch Fraunhofer-Publica (publica.fraunhofer.de)

Researcher

Publications (selection)