Projects

UnMute aims to address the limitations of today's speech and voice-based interactions and open up intelligent interfaces to the currently digitally ‘unheard’. In this project, I have developed methods for zero-resource cross-lingual transfer of ASR models and methods for spoken information retrieval in unwritten languages.

The goal of COG-MHEAR is to develop cognitively-inspired 5G-IoT enabled, multi-modal hearing aids. In this project, I helped co-organise the Audio-Visual Speech Enhancement challenge and I worked on devising new protocols for evaluation of speech enhancements models.

The goal of Material was to develop a tool that would allow analysts to consume foreign text and speech media regardless of their language expertise. I worked on semi-supervised training of ASR models that allowed us to train models with small amounts of manually transcribed data and large amounts of crawled text and speech.

The goal of SUMMA was to develop a platform to automate the analysis of media streams across many languages. In this project, I developed methods for automatic punctuation prediction in ASR transcripts and I was responsible deployment of ASR and punctuation models.

CloudASR is a cloud based automatic speech recognition platform which supports both batch and online speech recognition. The key features are scalability, customizability and easy deployment. It is tested to be able to handle more than 1000 parallel requests given enough computational resources.

MT–ComparEval is a tool for comparison and evaluation of machine translation. Translations can be compared according to automatic metrics for whole documents or single sentences. Differences between translations can be visualized by highlighting matching/missing n-grams etc.

Publications

2024

C. Minixhofer, O. Klejch, P. Bell "TTSDS — Text-to-Speech Distribution Score" In SLT 2024.

T. Reitmaier, E. Wallington, O. Klejch, D. K. Raju, N. Markl, E. Nielsen, G. Bailey, J. Pearson, M. Jones, P. Bell, S. Robinson "UnMute Toolkit: Speech Interactions Designed With Minoritised Language Speakers" In CUI 2024.

T. Reitmaier, D. K. Raju, O. Klejch, E. Wallington, N. Markl, J. Pearson, M. Jones, P. Bell, S. Robinson "Cultivating Spoken Language Technologies for Unwritten Languages" In CHI 2024 (honorable mention).

A. Hussein, D. Zeinali, O. Klejch, M. Wiesner, B. Yan, S. Chowdhury, A. Ali, S. Watanabe, S. Khudanpur "Speech collage: code-switched audio generation by collaging monolingual corpora" In ICASSP 2024.

2023

L.-M. Lam-Yee-Mui, L. Ondel, O. Klejch "Comparing Self-Supervised Pre-Training and Semi-Supervised Training for Speech Recognition in Languages with Weak Language Models" In Interspeech 2023.

C. Minixhofer, O. Klejch, P. Bell "Evaluating and reducing the distance between synthetic and real speech distributions" In Interspeech 2023.

R. Sanabria, O. Klejch, H. Tang, S. Goldwater "Acoustic Word Embeddings for Untranscribed Target Languages with Continued Pretraining and Learned Pooling" In Interspeech 2023.

Y. Li, Z. Zhao, O. Klejch, P. Bell, C. Lai "ASR and Emotional Speech: A Word-Level Investigation of the Mutual Impact of Speech and Emotion Recognition" In Interspeech 2023.

N. Markl, E. Wallington, O. Klejch, T. Reitmaier, G. Bailey, J. Pearson, M. Jones, S. Robinson, P. Bell "Automatic transcription and (de) standardisation" In SIGUL 2023.

R. Sanabria, N. Bogoychev, N. Markl, A. Carmantini, O. Klejch, P. Bell "The Edinburgh International Accents of English Corpus: Towards the Democratization of English ASR" In ICASSP 2023.

C. Valentini Botinhao, A. L. Aldana Blanco, O. Klejch, P. Bell "Efficient Inteligibility Evaluation using Keyword Spotting: A Study on Audio-Visual Speech Enhancement" In ICASSP 2023.

B. Yan, M. Wiesner, O. Klejch, P. Jyothi, S. Watanabe "Towards Zero-Shot Code-Switched Speech Recognition" In ICASSP 2023.

T. Reitmaier, E. Wallington, O. Klejch, N. Markl, L.-M. Lam-Yee-Mui, J. Pearson, M. Jones, P. Bell, S. Robinson "Situating Automatic Speech Recognition Development within Communities of Under-heard Language Speakers" In CHI 2023.

2022

A. L. Aldana Blanco, C. Valentini Botinhao, O. Klejch, M. Gogate, K. Dashtipour, A. Hussain, P. Bell "AVSE Challenge: Audio-Visual Speech Enhancement Challenge" In SLT 2022.

O. Klejch, E. Wallington, P.Bell "Deciphering Speech: a Zero-Resource Approach to Cross-Lingual Transfer in ASR." In Interspeech 2022.

T. Reitmaier, E. Wallington, D. K. Raju, O. Klejch, J. Pearson, M. Jones, P. Bell, and S. Robinson "Opportunities and Challenges of Automatic Speech Recognition Systems for Low-Resource Language Speakers." In CHI 2022.

2021

E. Wallington, B. Kershenbaum, O. Klejch, P. Bell "On the learning dynamics of semi-supervised training for ASR." In Interspeech 2021.

O. Klejch, E. Wallington, P. Bell "The CSTR System for Multilingual and Code-Switching ASR Challenges for Low Resource Indian Languages." In Interspeech 2021.

2020

P. Bell, J. Fainberg, O. Klejch, J. Li, S. Renals, P. Swietojanski "Adaptation algorithms for neural network-based speech recognition: An overview." In IEEE Open Journal of Signal Processing, 2020.

J. Roth, S. Chaudhuri, O. Klejch, R. Marvin, A. Gallagher, L. Kaver, S. Ramaswamy, A. Stopczynski, C. Schmid, Z. Xi, C. Pantofaru "Ava active speaker: An audio-visual dataset for active speaker detection." In ICASSP 2020.

2019

O. Klejch, J. Fainberg, P. Bell, S. Renals "Speaker adaptive training using model agnostic meta-learning." In ASRU 2019.

J. Fainberg, O. Klejch, E. Loweimi, P. Bell, S. Renals "Acoustic model adaptation from raw waveforms with SincNet." In ASRU 2019.

J. Fainberg, O. Klejch, P. Bell, S. Renals "Lattice-based lightly-supervised acoustic model training." In Interspeech 2019.

O. Klejch, J. Fainberg, P. Bell, S. Renals "Lattice-based unsupervised test-time adaptation of neural network acoustic models" arxiv:1906.11521, 2019.

2018

O. Klejch, J. Fainberg, P. Bell. "Learning to adapt: a meta-learning approach for speaker adaptation." In Interspeech 2018.

2017

E. Tsunoo, O. Klejch, P. Bell, S. Renals. "Hierarchical recurrent neural network for story segmentation using fusion of lexical and acoustic features." In ASRU 2017.

O. Klejch, P. Bell, and S. Renals. "Sequence-to-sequence models for punctuated transcription combining lexical and acoustic features." In ICASSP 2017.

2016

O. Klejch, P. Bell, and S. Renals. "Punctuated transcription of multi-genre broadcasts using acoustic and lexical approaches." In SLT 2016.

R. Sudarikov, M. Popel, O. Bojar, A. Burchardt, O. Klejch. "Using MT-ComparEval." In LREC 2016 Workshop: Translation Evaluation 2016.

N. Aranberri, E. Avramidis, A. Burchardt, O. Klejch, M. Popel and M. Popović. "Tools and Guidelines for Principled Machine Translation Development." In LREC 2016.

2015

O. Klejch, O. Platek, L. Zilka, and F. Jurcicek. "CloudASR: Platform and Service." In Text, Speech, and Dialogue, pp. 334-341. Springer International Publishing, 2015.

O. Klejch, E. Avramidis, A. Burchardt, and M. Popel. "MT-ComparEval: Graphical evaluation interface for Machine Translation development." The Prague Bulletin of Mathematical Linguistics 104, no. 1 (2015): 63-74.

2014

O. Klejch, O. Platek, L. Zilka, and F. Jurcicek. "CloudASR: Platform and Service." Demo in SLT 2014.