Erfan Loweimi
erfan loweimi

Dr Erfan Loweimi

  

Biography

Erfan Loweimi holds a part-time EPSRC Research Fellowship at School of Computing, Engineering & The Built Environment, and serves as a full-time Research Associate at the Machine Intelligence Laboratory, University of Cambridge (2022-). Prior to his current roles, he held Research Associate positions at King’s College London (2021-2023) and at the Centre for Speech Technology Research (CSTR), University of Edinburgh (2018-2021). He earned his PhD from the University of Sheffield in 2018, where he was a Faculty Scholar in the Speech and Hearing Research Group (SPandH), Department of Computer Science.

Erfan has received recognition for his contributions, including the “Research Communicator of the Year Award (University of Sheffield, 2017)” and the “Outstanding Reviewer Award (IEEE ICASSP, 2022).” He has actively served his community in various capacities, including as an Area Chair in prestigious conferences (INTERSPEECH, ICASSP, EMNLP), Publication Chair, Organising Committee Member in multiple conferences, and as an associate member of the IEEE’s Speech and Language Technical Committee (SLTC).

Dr. Loweimi has published 38 peer-reviewed papers and is the first author of more than 28 journal and conference papers, including three in IEEE Transactions on Audio, Speech, and Language Processing. His research interests encompass end-to-end speech processing and recognition, applications of speech technology in healthcare, explainable and trustworthy AI-based speech technology, multi-modal speech processing, and multi-modal information retrieval.

School

Research Groups

Esteem

Conference Organising Activity

  • Meta Reviewer in IEEE ICASSP 2024
  • Area Chair in ISCA INTERSPEECH 2024
  • UKISpeech 2024 Co-organiser
  • Area Chair in ISCA INTERSPEECH 2023
  • Area Chair in EMNLP 2023
  • Meta Reviewer in IEEE ICASSP 2023
  • Publication Chair in IEEE Spoken Language Technology Workshop (SLT)
  • UKSpeech 2016 co-organiser

 

Fellowships and Awards

  • Outstanding Reviewer Award (IEEE ICASSP 2022)
  • Research Communicator of the Year Award (University of Sheffield, 2017)

 

Invited Speaker

  • Speaker Retrieval in the Wild: Challenges, Effectiveness and Robustness, University of Cambridge, Cambridge, UK, 2024
  • Speaker Retrieval in the Wild, BBC Broadcasting House, London, UK, 2024
  • Phonetic Error Analysis beyond Phone Error Rate, University of Edinburgh, Edinburgh, UK, 2023
  • Recent Advances in Interpreting and Understanding DNNs, Iranian Conference on Machine Vision and Image Processing (MVIP), Iran, 2022
  • Speech Acoustic Modelling from Raw Signal Representations, Edinburgh Napier University, Edinburgh, UK, 2022
  • On the Robustness and Training Dynamics of Raw Waveform Models, University of Edinburgh, Edinburgh, UK, 2021
  • Raw Sign and Magnitude Spectra for Multi-head Acoustic Modelling, University of Edinburgh, Edinburgh, UK, 2020
  • DNN Statistical Interpretation and Normalisation for ASR, University of Edinburgh, Edinburgh, UK, 2019
  • Understanding and Interpreting DNNs for Speech Recognition, Qatar Computing Research Institute (QCRI), Doha, Qatar, 2019
  • Robust Phase-based Speech Signal Processing; From Source-Filter Separation to Model-Based Robust ASR, University of Sheffield, Sheffield, UK, 2018
  • Speech Phase Spectrum: Love It or Leave It?, University of Edinburgh, Edinburgh, UK, 2018
  • Genie in the mike! The Science of Talking (with) Machines, A Pint of Science Festival, Sheffield, 2017
  • Channel Compensation in the Generalised VTS Approach to Robust ASR, UKSpeech 2017, University of Cambridge, Cambridge, UK, 2017
  • Signal Processing is Dead(?)! Long Live DNN!, Machine Intelligence for Natural Interfaces (MINI) workshop, Sheffield, 2016
  • Deep Learning, The End of History and The Last Computer Scientist, A Pint of Science Festival, Sheffield, 2016
  • Source-filter Separation of Speech Signal in the Phase Domain, UKSpeech 2015, University of East Anglia, Norwich, UK, 2015

 

Membership of Professional Body

  • IEEE Signal Processing Society
  • International Speech Communication Association (ISCA)

 

Public/Community Engagement

  • Genie in the mike! The Science of Talking (with) Machines, A Pint of Science Festival, Sheffield, UK, 2017
  • Deep Learning, The End of History and The Last Computer Scientist, A Pint of Science Festival, Sheffield, UK, 2016

 

Reviewing

  • IEEE/ACM Transactions on Audio, Speech, and Language Processing
  • International Conference on Affective Computing & Intelligent Interaction (ACII)
  • IEEE Automatic Speech and Understanding Workshop (ASRU)
  • Computer Speech & Language
  • IEEE International Conference on Acoustics, Speech, and Signal Processing (ICASSP)
  • ISCA INTERSPEECH

 

Visiting Positions

  • Visitor in King's College London (KCL)
  • Visitor in Centre for Speech Technology Research (CSTR), University of Edinburgh

 

Filter by

Date


29 results

Speech Acoustic Modelling from Raw Phase Spectrum

Conference Proceeding
Loweimi, E., Cvetkovic, Z., Bell, P., & Renals, S. (2021)
Speech Acoustic Modelling from Raw Phase Spectrum. In ICASSP 2021 - 2021 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP). https://doi.org/10.1109/icassp39728.2021.9413727
Magnitude spectrum-based features are the most widely employed front-ends for acoustic modelling in automatic speech recognition (ASR) systems. In this paper, we investigate t...

Train Your Classifier First: Cascade Neural Networks Training from Upper Layers to Lower Layers

Conference Proceeding
Zhang, S., Do, C., Doddipatla, R., Loweimi, E., Bell, P., & Renals, S. (2021)
Train Your Classifier First: Cascade Neural Networks Training from Upper Layers to Lower Layers. In ICASSP 2021 - 2021 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP). https://doi.org/10.1109/icassp39728.2021.9413565
Although the lower layers of a deep neural network learn features which are transferable across datasets, these layers are not transferable within the same dataset. That is, i...

On The Usefulness of Self-Attention for Automatic Speech Recognition with Transformers

Conference Proceeding
Zhang, S., Loweimi, E., Bell, P., & Renals, S. (2021)
On The Usefulness of Self-Attention for Automatic Speech Recognition with Transformers. In 2021 IEEE Spoken Language Technology Workshop (SLT). https://doi.org/10.1109/slt48900.2021.9383521
Self-attention models such as Transformers, which can capture temporal relationships without being limited by the distance between events, have given competitive speech recogn...

Raw Sign and Magnitude Spectra for Multi-Head Acoustic Modelling

Conference Proceeding
Loweimi, E., Bell, P., & Renals, S. (2020)
Raw Sign and Magnitude Spectra for Multi-Head Acoustic Modelling. In Proc. Interspeech 2020 (1644-1648). https://doi.org/10.21437/interspeech.2020-18
In this paper we investigate the usefulness of the sign spectrum and its combination with the raw magnitude spectrum in acoustic modelling for automatic speech recognition (AS...

On the Robustness and Training Dynamics of Raw Waveform Models

Conference Proceeding
Loweimi, E., Bell, P., & Renals, S. (2020)
On the Robustness and Training Dynamics of Raw Waveform Models. In Proc. Interspeech 2020 (1001-1005). https://doi.org/10.21437/interspeech.2020-17
We investigate the robustness and training dynamics of raw waveform acoustic models for automatic speech recognition (ASR). It is known that the first layer of such models lea...

Acoustic Model Adaptation from Raw Waveforms with Sincnet

Conference Proceeding
Fainberg, J., Klejch, O., Loweimi, E., Bell, P., & Renals, S. (2019)
Acoustic Model Adaptation from Raw Waveforms with Sincnet. In 2019 IEEE Automatic Speech Recognition and Understanding Workshop (ASRU). https://doi.org/10.1109/asru46091.2019.9003974
Raw waveform acoustic modelling has recently gained interest due to neural networks' ability to learn feature extraction, and the potential for finding better representations ...

On Learning Interpretable CNNs with Parametric Modulated Kernel-Based Filters

Conference Proceeding
Loweimi, E., Bell, P., & Renals, S. (2019)
On Learning Interpretable CNNs with Parametric Modulated Kernel-Based Filters. In Proc. Interspeech 2019 (3480-3484). https://doi.org/10.21437/interspeech.2019-1257
We investigate the problem of direct waveform modelling using parametric kernel-based filters in a convolutional neural network (CNN) framework, building on SincNet, a CNN emp...

Learning Temporal Clusters Using Capsule Routing for Speech Emotion Recognition

Conference Proceeding
Jalal, M. A., Loweimi, E., Moore, R. K., & Hain, T. (2019)
Learning Temporal Clusters Using Capsule Routing for Speech Emotion Recognition. In Proc. Interspeech 2019 (1701-1705). https://doi.org/10.21437/interspeech.2019-3068
Emotion recognition from speech plays a significant role in adding emotional intelligence to machines and making human-machine interaction more natural. One of the key challen...

Trainable Dynamic Subsampling for End-to-End Speech Recognition

Conference Proceeding
Zhang, S., Loweimi, E., Xu, Y., Bell, P., & Renals, S. (2019)
Trainable Dynamic Subsampling for End-to-End Speech Recognition. In Proc. Interspeech 2019 (1413-1417). https://doi.org/10.21437/interspeech.2019-2778
Jointly optimised attention-based encoder-decoder models have yielded impressive speech recognition results. The recurrent neural network (RNN) encoder is a key component in s...

On the Usefulness of Statistical Normalisation of Bottleneck Features for Speech Recognition

Conference Proceeding
Loweimi, E., Bell, P., & Renals, S. (2019)
On the Usefulness of Statistical Normalisation of Bottleneck Features for Speech Recognition. In ICASSP 2019 - 2019 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP). https://doi.org/10.1109/icassp.2019.8683330
DNNs play a major role in the state-of-the-art ASR systems. They can be used for extracting features and building probabilistic models for acoustic and language modelling. Des...