Professor Thomas Hain
Department of Computer Science
Professor of Speech and Audio Technology
Director of UKRI CDT in Speech and Language Technologies
Director of Centre for Speech and Language Technology (VoiceBase)
Head of the Speech and Hearing (SpandH) research group
t.hain@sheffield.ac.uk
+44 114 222 1836
+44 114 222 1836
Regent Court (DCS)
Full contact details
Professor Thomas Hain
Department of Computer Science
Regent Court (DCS)
211 Portobello
Sheffield
S1 4DP
Department of Computer Science
Regent Court (DCS)
211 Portobello
Sheffield
S1 4DP
- Profile
-
Thomas Hain obtained the degree 'Dipl.-Ing' in Electrical/Communication Engineering in 1994 from the University of Technology, Vienna. He joined the Speech Technology Group at Philips Speech Processing which he left in a senior position.
In 1997 he joined the Speech, Vision and Robotics Group at the Cambridge University Engineering Department as Research Associate and PhD Student. He took up a Lectureship at the SVR group in 2001.
In 2004 he joined the Speech and Hearing Group to work as Lecturer in Computer Science. He was promoted to Senior Lecturer in 2008 and Reader in 2011.
- Research interests
-
Thomas' research interests cover many areas in natural language processing, speech, audio and multimedia technology, machine learning, and complex system optimisation and design.
His interests include: large vocabulary continuous speech recognition, non-linear methods in speech processing, low bit-rate speech coding, machine learning, multi-modal systems, image classification, microphone arrays, system and resource optimisation.
- Publications
-
Books
- The HTK Book. Cambridge, England: Cambridge University Engineering Department.
- The HTK Book (for HTK Version 3.5, documentation alpha version). Cambridge University Engineering Department: Cambridge University Engineering Department.
Journal articles
- Automatic detection of behavioural codes in team interactions. Computer Speech and Language, 74.
- Att-TasNet: attending to encodings in time-domain audio speech separation of noisy, reverberant speech mixtures. Frontiers in Signal Processing, 2. View this article in WRRO
- H-VECTORS: Improving the robustness in utterance-level speaker embeddings using a hierarchical attention model.. Neural Netw, 142, 329-339.
- Evaluation of the effectiveness and efficiency of state-of-the-art features and models for automatic speech recognition error detection. Journal of Big Data, 8. View this article in WRRO
- System-Independent ASR error detection and classification using Recurrent Neural Network. Computer Speech & Language, 55, 187-199. View this article in WRRO
- Recurrent Neural Network Language Model Adaptation for Multi-Genre Broadcast Speech Recognition and Alignment. IEEE/ACM Transactions on Audio, Speech and Language Processing, 27(3), 572-582. View this article in WRRO
- Lightly supervised alignment of subtitles on multi-genre broadcasts. Multimedia Tools and Applications, 77(23), 30533-30550. View this article in WRRO
- Unsupervised crosslingual adaptation of tokenisers for spoken language recognition. Computer Speech and Language, 46, 327-342. View this article in WRRO
- Acoustic adaptation to dynamic background conditions with asynchronous transformations. Computer Speech and Language, 41, 180-194. View this article in WRRO
- Capitalising on North American speech resources for the development of a South African English large vocabulary speech recognition system. Computer Speech and Language, 28(6), 1255-1268.
- Lightly supervised learning from a damaged natural speech corpus. ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings, 8086-8090.
- Correctness-adjusted unsupervised discriminative acoustic model adaptation. IEEE Transactions on Audio, Speech and Language Processing, PP(99).
- Introduction to the Special Section on New Frontiers in Rich Transcription. IEEE TRANSACTIONS ON AUDIO SPEECH AND LANGUAGE PROCESSING, 20(2), 353-355.
- Supervised and unsupervised Web-based language model domain adaptation. 13th Annual Conference of the International Speech Communication Association 2012, INTERSPEECH 2012, 1, 182-185.
- Automatic transcription of academic lectures from diverse disciplines. 2012 IEEE Workshop on Spoken Language Technology, SLT 2012 - Proceedings, 398-403.
- Application of SVM-based correctness predictions to unsupervised discriminative speaker adaptation. ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings, 4341-4344.
- Transcribing meetings with the AMIDA systems. IEEE Transactions on Audio, Speech and Language Processing.
- Error approximation and minimum phone error acoustic model estimation. IEEE Transactions on Audio, Speech and Language Processing, 18(6), 1269-1279.
- Automatic Optimization of Speech Decoder Parameters. IEEE SIGNAL PROC LET, 17(1), 95-98.
- Discrimininative training of narrow band - Wide band adapted systems for meeting recognition. Proceedings of the Annual Conference of the International Speech Communication Association, INTERSPEECH, 1217-1220.
- Automatic speech recognition for scientific purposes - WebASR. Proceedings of the Annual Conference of the International Speech Communication Association, INTERSPEECH, 504-507.
- Discrimininative training of narrow band - Wide band adapted systems for meeting recognition. INTERSPEECH 2008 - 9th Annual Conference of the International Speech Communication Association, 1217-1220.
- Application of CMLLR in narrow band wide band adapted systems. International Speech Communication Association - 8th Annual Conference of the International Speech Communication Association, Interspeech 2007, 4, 2860-2863.
- Recognition and understanding of meetings the AMI and AMIDA projects. 2007 IEEE Workshop on Automatic Speech Recognition and Understanding, ASRU 2007, Proceedings, 238-247.
- The AMI system for the transcription of speech in meetings. ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings, 4.
- Corrections to "Automatic Transcription of Conversational Telephone Speech".. IEEE Trans. Speech Audio Process., 14, 727-727.
- Strategies for language model web-data collection. ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings, 1.
- Automatic transcription of conversational telephone speech. IEEE T SPEECH AUDI P, 13(6), 1173-1185.
- Implicit modelling of pronunciation variation in automatic speech recognition. SPEECH COMMUNICATION, 46(2), 171-188.
Chapters
- Use of Speaker Metadata for Improving Automatic Pronunciation Assessment, Statistical Language and Speech Processing (pp. 61-72).
- Speech recognition, Multimodal Signal Processing (pp. 56-83). Cambridge University Press
- Speech Recognition In Clark A, Fox C & Lappin S (Ed.), The Handbook of Computational Linguistics and Natural Language Processing (pp. 299-332). Wiley-Blackwell
- Juicer: A weighted finite-state transducer speech decoder (pp. 285-296).
- The AMI Meeting Corpus: A Pre-announcement, Machine Learning for Multimodal Interaction, Lecture Notes in Computer Science (pp. 28-39). Edinburgh: Springer.
Conference proceedings papers
- Non-intrusive Speech Intelligibility Estimated By Metric Prediction for Hearing Impaired Individuals for the Clarity Prediction Challenge 1. Proceedings of the Annual Conference of the International Speech Communication Association, INTERSPEECH, 18 September 2022 - 22 September 2022.
- Unsupervised Data Selection for Speech Recognition with Contrastive Loss Ratios. ICASSP 2022 - 2022 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), 23 May 2022 - 27 May 2022.
- A Model for Assessor Bias in Automatic Pronunciation Assessment. ICASSP 2022 - 2022 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP) (pp 7267-7271)
- MetricGAN+/-: Increasing Robustness of Noise Reduction on Unseen Data. Proc. 30th European Signal Processing Conference, EUSIPCO 2022. Belgrade, Serbia, 29 August 2022 - 2 September 2022.
- Attention Based Model for Segmental Pronunciation Error Detection. 2021 IEEE Automatic Speech Recognition and Understanding Workshop (ASRU), 13 December 2021 - 17 December 2021.
- Towards Low-Resource Stargan Voice Conversion Using Weight Adaptive Instance Normalization. ICASSP 2021 - 2021 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), 6 June 2021 - 11 June 2021.
- Multiple-Hypothesis CTC-Based Semi-Supervised Adaptation of End-to-End Speech Recognition. ICASSP 2021 - 2021 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), 6 June 2021 - 11 June 2021.
- Improving audio anomalies recognition using temporal convolutional attention networks. ICASSP 2021 - 2021 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP) (pp 6473-6477). Toronto, ON, Canada, 6 June 2021 - 11 June 2021.
- Supervised speaker embedding de-mixing in two-speaker environment. 2021 IEEE Spoken Language Technology Workshop (SLT) (pp 758-765). Shenzhen, China, 19 January 2021 - 22 January 2021.
- Selective Adaptation of End-to-End Speech Recognition using Hybrid CTC/Attention Architecture for Noise Robustness. 2020 28th European Signal Processing Conference (EUSIPCO), 18 January 2021 - 21 January 2021.
- Contextual Joint Factor Acoustic Embeddings, Vol. 00 (pp 750-757)
- WINVC: One-Shot Voice Conversion with Weight Adaptive Instance Normalization (pp 559-573)
- Uncertainty Aware Review Hallucination for Science Article Classification. Findings of the Association for Computational Linguistics: ACL-IJCNLP 2021 (pp 5004-5009)
- Robust speaker recognition using speech enhancement and attention model. The Speaker and Language Recognition Workshop (Odyssey 2020) (pp 451-458). Tokyo, Japan, 1 November 2020 - 5 November 2020.
- Removing Bias with Residual Mixture of Multi-View Attention for Speech Emotion Recognition. Interspeech 2020 (pp 4084-4088). Shanghai, China, 25 October 2020 - 29 October 2020.
- Weakly supervised training of hierarchical attention networks for speaker identification. Proceedings of Interspeech 2020 (pp 2992-2996). Shanghai, China, 25 October 2020 - 29 October 2020.
- Exploration of audio quality assessment and anomaly localisation using attention models. Proceedings of Interspeech 2020 (pp 4611-4615). Shanghai, China, 25 October 2020 - 29 October 2020.
- Speaker re-identification with speaker dependent speech enhancement. Proceedings of Interspeech 2020 (pp 1530-1534). Shanghai, China, 25 October 2020 - 29 October 2020.
- H-vectors : utterance-level speaker embedding using a hierarchical attention model. ICASSP 2020 - 2020 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP) (pp 7579-7583). Barcelona, Spain (virtual), 4 May 2020 - 8 May 2020.
- Unsupervised Adaptation of Acoustic Models for ASR Using Utterance-Level Embeddings from Squeeze and Excitation Networks. 2019 IEEE Automatic Speech Recognition and Understanding Workshop (ASRU), 14 December 2019 - 18 December 2019.
- Spatio-Temporal Context Modelling for Speech Emotion Classification. 2019 IEEE Automatic Speech Recognition and Understanding Workshop (ASRU), 14 December 2019 - 18 December 2019.
- A Cross-Corpus Study on Speech Emotion Recognition. 2019 IEEE Automatic Speech Recognition and Understanding Workshop (ASRU), 14 December 2019 - 18 December 2019.
- Message from the technical program chairs. Proceedings of the Annual Conference of the International Speech Communication Association, INTERSPEECH, Vol. 2019-September (pp 13-15)
- Improving ASR Error Detection with RNNLM Adaptation. 2018 IEEE Spoken Language Technology Workshop (SLT), 18 December 2018 - 21 December 2018.
- Exploring the use of group delay for generalised VTS based noise compensation. 2018 IEEE International Conference on Acoustics, Speech and Signal Processing Proceedings, 15 April 2018 - 20 April 2018. View this article in WRRO
- Improved acoustic modelling for automatic literacy assessment of children. Proceedings of Interspeech 2018 (pp 1666-1670), 2 September 2018 - 6 September 2018. View this article in WRRO
- Towards a generic approach for automatic speech recognition error detection and classification. 2018 4th International Conference on Advanced Technologies for Signal and Image Processing (ATSIP), 21 March 2018 - 24 March 2018.
- Exploring the use of acoustic embeddings in neural machine translation. 2017 IEEE Automatic Speech Recognition and Understanding Workshop (ASRU), 16 December 2017 - 20 December 2017. View this article in WRRO
- Semi-supervised Adaptation of RNNLMs by Fine-tuning with Domain-specific Auxiliary Features. Proceedings of the Annual Conference of the International Speech Communication Association, INTERSPEECH (pp 2715-2719) View this article in WRRO
- Robust Source-Filter Separation of Speech Signal in the Phase Domain. Proceedings of the Annual Conference of the International Speech Communication Association View this article in WRRO
- DNN approach to speaker diarisation using speaker channels. Proceedings - ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing (pp 4925-4929) View this article in WRRO
- Analysing Acoustic Model Changes for Active Learning in Automatic Speech Recognition. International Conference on Systems, Signals and Image Processing (IWSSIP) View this article in WRRO
- Shefce: A Cantonese-English bilingual speech corpus for pronunciation assessment. 2017 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), 5 March 2017 - 9 March 2017. View this article in WRRO
- Statistical normalisation of phase-based feature representation for robust speech recognition. ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings (pp 5310-5314) View this article in WRRO
- Automatic speech recognition errors detection using supervised learning techniques. 2016 IEEE/ACS 13th International Conference of Computer Systems and Applications (AICCSA), 29 November 2016 - 2 December 2016. View this article in WRRO
- The 2015 Sheffield system for transcription of Multi-Genre Broadcast media. 2015 IEEE Workshop on Automatic Speech Recognition and Understanding, ASRU 2015 - Proceedings (pp 624-631) View this article in WRRO
- Error correction in lightly supervised alignment of broadcast subtitles. Proceedings of the 17th Annual Conference of the International Speech Communication Association (Interspeech) View this article in WRRO
- webASR 2 - Improved cloud based speech technology. Proceedings of the 17th Annual Conference of the International Speech Communication Association (Interspeech) (pp 1613-1617) View this article in WRRO
- Using phone features to improve dialogue state tracking
generalisation to unseen states. Proceedings of the 17th Annual Meeting of the Special Interest Group
on Discourse and Dialogue, September 2016 - September 2016. View this article in WRRO
- Use of generalised nonlinearity in Vector Taylor Series noise compensation for robust speech recognition. Proceedings of the Annual Conference of the International Speech Communication Association, INTERSPEECH, Vol. 08-12-September-2016 (pp 3798-3802) View this article in WRRO
- Combining weak tokenisers for phonotactic language recognition in a resource-constrained setting. Combining weak tokenisers for phonotactic language recognition in a resource-constrained setting (pp 2939-2943), 9 September 2016 - 12 September 2016. View this article in WRRO
- Emotion Recognition from the Speech Signal by Effective Combination of Generative and Discriminative Models. USES
- The Sheffield language recognition system in NIST LRE 2015. Proceedings of The Speaker and Language Recognition Workshop Odyssey 2016 View this article in WRRO
- A framework for collecting realistic recordings of dysarthric speech - The homeService corpus. Proceedings of the 10th International Conference on Language Resources and Evaluation, LREC 2016 (pp 1993-1997) View this article in WRRO
- Groupwise learning for ASR k-best list reranking in spoken language translation. ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings, Vol. 2016-May (pp 6120-6124) View this article in WRRO
- Segment-oriented evaluation of speaker diarisation performance. 2016 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), 20 March 2016 - 25 March 2016. View this article in WRRO
- The OpenCourseWare metadiscourse (OCWMD) corpus. Proceedings of the 10th International Conference on Language Resources and Evaluation, LREC 2016 (pp 1770-1776)
- The 2015 sheffield system for longitudinal diarisation of broadcast media. 2015 IEEE Workshop on Automatic Speech Recognition and Understanding (ASRU), 13 December 2015 - 17 December 2015. View this article in WRRO
- Source-filter separation of speech signal in the phase domain. Proceedings of the Annual Conference of the International Speech Communication Association, INTERSPEECH, Vol. 2015-January (pp 598-602) View this article in WRRO
- Unsupervised domain discovery using latent dirichlet allocation for acoustic modelling in speech recognition. Proceedings of the Annual Conference of the International Speech Communication Association, INTERSPEECH, Vol. 2015-January (pp 3640-3644) View this article in WRRO
- Latent Dirichlet Allocation based organisation of broadcast media archives for deep neural network adaptation. 2015 IEEE Workshop on Automatic Speech Recognition and Understanding (ASRU), 13 December 2015 - 17 December 2015. View this article in WRRO
- The MGB challenge: Evaluating multi-genre broadcast media recognition. 2015 IEEE Workshop on Automatic Speech Recognition and Understanding (ASRU), 13 December 2015 - 17 December 2015. View this article in WRRO
- Knowledge transfer between speakers for personalised dialogue management. Proceedings of the 16th Annual Meeting of the Special Interest Group on Discourse and Dialogue, September 2015 - September 2015.
- Data-selective transfer learning for multi-domain speech recognition. Proceedings of the Annual Conference of the International Speech Communication Association, INTERSPEECH, Vol. 2015-January (pp 2897-2901) View this article in WRRO
- An investigation into speaker informed DNN front-end for LVCSR. 2015 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), 19 April 2015 - 24 April 2015. View this article in WRRO
- Automatic assessment of English learner pronunciation using discriminative classifiers. 2015 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), 19 April 2015 - 24 April 2015. View this article in WRRO
- AN INVESTIGATION INTO SPEAKER INFORMED DNN FRONT-END FOR LVCSR. 2015 IEEE INTERNATIONAL CONFERENCE ON ACOUSTICS, SPEECH, AND SIGNAL PROCESSING (ICASSP) (pp 4300-4304)
- QUALITY ESTIMATION FOR ASR K-BEST LIST RESCORING IN SPOKEN LANGUAGE TRANSLATION. 2015 IEEE INTERNATIONAL CONFERENCE ON ACOUSTICS, SPEECH, AND SIGNAL PROCESSING (ICASSP) (pp 5226-5230)
- Annotating meta-discourse in academic lectures from different disciplines.. SLaTE (pp 161-166)
- The 2015 sheffield system for transcription of Multi-Genre Broadcast media.. ASRU (pp 624-631)
- Speech-enabled environmental control in an AAL setting for people with speech disorders: a case study. IET International Conference on Technologies for Active and Assisted Living (TechAAL)
- Noise-matched training of CRF based sentence end detection models. Proceedings of the Annual Conference of the International Speech Communication Association, INTERSPEECH, Vol. 2015-January (pp 349-353)
- A study on the stability and effectiveness of features in quality estimation for spoken language translation. Proceedings of the Annual Conference of the International Speech Communication Association, INTERSPEECH, Vol. 2015-January (pp 2257-2261)
- Long-Term Statistical Feature Extraction from Speech Signal and Its Application in Emotion Recognition (pp 173-184)
- Quality estimation for asr k-best list rescoring in spoken language translation. 2015 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), 19 April 2015 - 24 April 2015.
- Using Topic Segmentation Models for the Automatic Organisation of MOOCs resources.. EDM (pp 524-527)
- Semi-supervised DNN training in meeting recognition. 2014 IEEE Spoken Language Technology Workshop (SLT), 7 December 2014 - 10 December 2014. View this article in WRRO
- Using neural network front-ends on far field multiple microphones based speech recognition. 2014 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP) View this article in WRRO
- Semi-supervised DNN training in meeting recognition. 2014 IEEE Workshop on Spoken Language Technology, SLT 2014 - Proceedings (pp 141-146)
- Background-tracking acoustic features for genre identification of broadcast shows. 2014 IEEE Spoken Language Technology Workshop (SLT), 7 December 2014 - 10 December 2014.
- Automatic selection of speakers for improved acoustic modelling: recognition of disordered speech with sparse data. 2014 IEEE Spoken Language Technology Workshop (SLT), 7 December 2014 - 10 December 2014.
- The USFD SLT System for IWSLT 2014. IWSLT View this article in WRRO
- Extending Limabeam with discrimination and coarse gradients. Proceedings of the Annual Conference of the International Speech Communication Association, INTERSPEECH (pp 2440-2444)
- Multi-pass sentence-end detection of lecture speech. Proceedings of the Annual Conference of the International Speech Communication Association, INTERSPEECH (pp 2902-2906)
- Speaker dependent bottleneck layer training for Speaker adaptation in automatic speech recognition. Proceedings of the Annual Conference of the International Speech Communication Association, INTERSPEECH (pp 2199-2203)
- Adaptive speech recognition and dialogue management for users with speech disorders. Proceedings of the Annual Conference of the International Speech Communication Association, INTERSPEECH (pp 1033-1037)
- Using contextual information in joint factor eigenspace MLLR for speech recognition in diverse scenarios. 2014 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), 4 May 2014 - 9 May 2014. View this article in WRRO
- Asynchronous factorisation of speaker and background with feature transforms in speech recognition. Proceedings of the 14th Annual Conference of the International Speech Communication Association (Interspeech) (pp 1238-1242), 25 August 2013 - 29 August 2013. View this article in WRRO
- Adaptation of lecture speech recognition system with machine translation output. Proceedings of the 38th IEEE International Conference on Acoustics, Speech, and Signal Processing (ICASSP). Vancouver, Canada
- The Sheffield Wargames Corpus. 14TH ANNUAL CONFERENCE OF THE INTERNATIONAL SPEECH COMMUNICATION ASSOCIATION (INTERSPEECH 2013), VOLS 1-5 (pp 1115-1119)
- Asynchronous Factorisation of Speaker and Background with Feature Transforms in Speech Recognition. 14TH ANNUAL CONFERENCE OF THE INTERNATIONAL SPEECH COMMUNICATION ASSOCIATION (INTERSPEECH 2013), VOLS 1-5 (pp 1237-1241)
- Learning speaker-specific pronunciations of disordered speech. 14TH ANNUAL CONFERENCE OF THE INTERNATIONAL SPEECH COMMUNICATION ASSOCIATION (INTERSPEECH 2013), VOLS 1-5 (pp 1158-1162)
- Combining in-domain and out-of-domain speech data for automatic recognition of disordered speech. 14TH ANNUAL CONFERENCE OF THE INTERNATIONAL SPEECH COMMUNICATION ASSOCIATION (INTERSPEECH 2013), VOLS 1-5 (pp 3609-3612)
- Automatic transcription of multi-genre media archives. CEUR Workshop Proceedings, Vol. 1012 (pp 26-31) View this article in WRRO
- Learning speaker-specific pronunciations of disordered speech. Proceedings of the Annual Conference of the International Speech Communication Association, INTERSPEECH (pp 1159-1163)
- Combining in-domain and out-of-domain speech data for automatic recognition of disordered speech. Proceedings of the Annual Conference of the International Speech Communication Association, INTERSPEECH (pp 3642-3645)
- The Sheffield Wargames Corpus.. 14th Annual Conference of the International Speech Communication Association (Interspeech 2013). Lyon, France, 25 August 2013 - 29 August 2013. View this article in WRRO
- HomeService: Voice-enabled assistive technology in the home using cloud-based automatic speech recognition. SLPAT 2013 - 4th Workshop on Speech and Language Processing for Assistive Technologies, SLPAT 2013, Workshop Proceedings (pp 29-34)
- CRF-based diacritisation of colloquial Arabic for automatic speech recognition. 13th Annual Conference of the International Speech Communication Association 2012, INTERSPEECH 2012, Vol. 3 (pp 1822-1825)
- An alignment matching method to explore pseudosyllable properties across different corpora. 13th Annual Conference of the International Speech Communication Association 2012, INTERSPEECH 2012, Vol. 1 (pp 862-865)
- A comparative study of adaptive, automatic recognition of disordered speech. 13th Annual Conference of the International Speech Communication Association 2012, INTERSPEECH 2012, Vol. 2 (pp 1774-1777)
- RESOURCE DEVELOPMENT AND EXPERIMENTS IN AUTOMATIC SOUTH AFRICAN BROADCAST NEWS TRANSCRIPTION. 3rd Workshop on Spoken Language Technologies for Under-Resourced Languages, SLTU 2012 (pp 102-106)
- Extending Audio Notetaker to Browse WebASR Transcriptions. Interspeech’11
- An Analysis of Automatic Speech Recognition with Multiple Microphones. Interspeech’11. Florence
- Web-based automatic speech recognition service - webASR. Interspeech’11
- An Investigation in Speech Recognition for Colloquial Arabic. Interspeech’11
- Making an automatic speech recognition service freely available on the web. Interspeech’11
- Cross-language phone recognition when the target language phoneme inventory is not known. Interspeech’11. Florence
- Extending Audio Notetaker to Browse WebASR Transcriptions. 12TH ANNUAL CONFERENCE OF THE INTERNATIONAL SPEECH COMMUNICATION ASSOCIATION 2011 (INTERSPEECH 2011), VOLS 1-5 (pp 3336-+)
- Meeting Recognition. Tutorial interspeech 2010
- The AMIDA 2009 Meeting Transcription System. 11TH ANNUAL CONFERENCE OF THE INTERNATIONAL SPEECH COMMUNICATION ASSOCIATION 2010 (INTERSPEECH 2010), VOLS 1-4 (pp 358-361)
- The AMIDA 2009 Meeting Transcription System. Interspeech’10 (pp 358-361)
- Real-Time ASR from Meetings. INTERSPEECH 2009: 10TH ANNUAL CONFERENCE OF THE INTERNATIONAL SPEECH COMMUNICATION ASSOCIATION 2009, VOLS 1-5 (pp 2067-+)
- Real-time ASR from meetings. Proceedings of the Annual Conference of the International Speech Communication Association, INTERSPEECH (pp 2119-2122)
- Interpretation of multiparty meetings the AMI and AMIDA projects. 2008 HANDS-FREE SPEECH COMMUNICATION AND MICROPHONE ARRAYS (pp 116-+)
- Interpretation of multiparty meetings the AMI and AMIDA projects. 2008 Hands-free Speech Communication and Microphone Arrays, Proceedings, HSCMA 2008 (pp 115-118)
- Automatic speech recognition for scientific purposes - webASR. INTERSPEECH 2008: 9TH ANNUAL CONFERENCE OF THE INTERNATIONAL SPEECH COMMUNICATION ASSOCIATION 2008, VOLS 1-5 (pp 504-507)
- BOB: A LEXICON AND PRONUNCIATION DICTIONARY GENERATOR. 2008 IEEE WORKSHOP ON SPOKEN LANGUAGE TECHNOLOGY: SLT 2008, PROCEEDINGS (pp 217-220)
- The 2007 AMI(DA) system for meeting transcription. MULTIMODAL TECHNOLOGIES FOR PERCEPTION OF HUMANS, Vol. 4625 (pp 414-428)
- Application of CMLLR in narrow band wide band adapted systems. Interspeech’07 (pp 282-285). Antwerp
- Temporal Masking for Unsupervised Minimum Bayes Risk Speaker Adaptation. INTERSPEECH 2007: 8TH ANNUAL CONFERENCE OF THE INTERNATIONAL SPEECH COMMUNICATION ASSOCIATION, VOLS 1-4 (pp 1577-1580)
- The AMI system for the transcription of speech in meetings. 2007 IEEE International Conference on Acoustics, Speech, and Signal Processing, Vol IV, Pts 1-3 (pp 357-360)
- The segmentation of multi-channel meeting recordings for automatic speech recognition. INTERSPEECH 2006 and 9th International Conference on Spoken Language Processing, INTERSPEECH 2006 - ICSLP, Vol. 3 (pp 1213-1216)
- Hypothesis Spaces For Minimum Bayes Risk Training In Large Vocabulary Speech Recognition. INTERSPEECH 2006 AND 9TH INTERNATIONAL CONFERENCE ON SPOKEN LANGUAGE PROCESSING, VOLS 1-5 (pp 2406-2409)
- Automatic Speech Recognition Experiments with Articulatory Data. INTERSPEECH 2006 AND 9TH INTERNATIONAL CONFERENCE ON SPOKEN LANGUAGE PROCESSING, VOLS 1-5 (pp 353-356)
- Audio-visual processing in meetings: Seven questions and current AMI answers. Machine Learning for Multimodal Interaction, Vol. 4299 (pp 24-35)
- Strategies for language model web-data collection. 2006 IEEE International Conference on Acoustics, Speech, and Signal Processing, Vol I, Proceedings (pp 1069-1072). Toulouse, FRANCE, 14 May 2006 - 19 May 2006.
- Strategies for language model web-data collection. 2006 IEEE INTERNATIONAL CONFERENCE ON ACOUSTICS, SPEECH AND SIGNAL PROCESSING, VOLS 1-13 (pp 1069-1072)
- The AMI meeting transcription system: Progress and performance. Machine Learning for Multimodal Interaction, Vol. 4299 (pp 419-431)
- The AMI Meeting Corpus. 5th International Conference on Methods and Techniques in Behavioral Research
- Applying vocal tract length normalization to meeting recordings. 9th European Conference on Speech Communication and Technology (pp 265-268)
- Transcription of conference room meetings: An investigation. 9th European Conference on Speech Communication and Technology (pp 1661-1664)
- The development of the AMI system for the transcription of speech in meetings. MACHINE LEARNING FOR MULTIMODAL INTERACTION, Vol. 3869 (pp 344-356)
- The 2005 AMI system for the transcription of speech in meetings. MACHINE LEARNING FOR MULTIMODAL INTERACTION, Vol. 3869 (pp 450-462)
- Progress in Broadcast News English Transcription. EARS STT Technical Meeting 2004. Montreal, Canada
- Cambridge STT Overview. EARS Mid-year Meeting 2004
- Using VTLN for Broadcast News Transcription. ICSLP’04. Cambridge University, UK
- Development of the 2003 CU-HTK conversational telephone speech transcription system. ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings, Vol. 1
- Development of the 2003 CU-HTK Conversational Telephone Speech transcription system. 2004 IEEE INTERNATIONAL CONFERENCE ON ACOUSTICS, SPEECH, AND SIGNAL PROCESSING, VOL I, PROCEEDINGS (pp 249-252)
- Single Pronunciation Dictionaries - Construction and Performance. EARS STT Technical Meeting 2004
- 2003 CU-HTK Broadcast News English System Development. Rich Transcription Workshop 2003s
- 2003 CU-HTK English CTS Systems. Rich Transcription Workshop 2003s. Boston, Ma
- CU-HTK RT-03 Mandarin CTS System. Rich Transcription Workshop 2003
- Recent Experiments with HTK Broadcast News and Conversational Telephone Systems. EARS Mid-year meeting 2003
- Recent advances in broadcast news transcription. ASRU'03: 2003 IEEE WORKSHOP ON AUTOMATIC SPEECH RECOGNITION AND UNDERSTANDING ASRU '03 (pp 105-110)
- Implicit Pronunciation Modelling in ASR. ITRW PMLA 2002. Estes Park, Colorado
- CU-HTK APRIL 2002 SWITCHBOARD SYSTEM. Rich Transcription Workshop 2002. Vienna, VA
- New features in the CU-HTK system for transcription of conversational telephone speech. 2001 IEEE INTERNATIONAL CONFERENCE ON ACOUSTICS, SPEECH, AND SIGNAL PROCESSING, VOLS I-VI, PROCEEDINGS (pp 57-60)
- The CU-HTK March 2000 HUB5E Transcription System. Speech Transcription Workshop 2000. College Park, Maryland
- Modelling sub-phone insertions and deletions in continuous speech recognition. ICSLP 2000
- Dynamic HMM selection for continuous speech recognition. Eurospeech’99 (pp 1327-1330). Budapest
- The 1998 HTK Broadcast News Transcription System: Development and Results. 1999 DARPA Broadcast News Transcription and Understanding Workshop. Herndon, VA
- The CUHTK-Entropic 10xRT Broadcast News Transcription System. 1999 DARPA Broadcast News Transcription and Understanding Workshop (pp 271-275). Herndon, VA
- Improvements in Accuracy and Speed in the HTK Broadcast News Transcription System. Eurospeech’99
- Hidden model sequences. Hub5 Workshop’99
- RECENT EXPERIMENTS WITH THE CU-HTK HUB5 SYSTEM. Hub5 Workshop’99
- The 1998 HTK system for transcription of conversational telephone speech. ICASSP '99: 1999 IEEE INTERNATIONAL CONFERENCE ON ACOUSTICS, SPEECH, AND SIGNAL PROCESSING, PROCEEDINGS VOLS I-VI (pp 57-60)
- SEGMENTATION AND CLASSIFICATION OF BROADCAST NEWS AUDIO. ICSLP’98
- The 1997 HTK Broadcast News Transcription System. 1998 DARPA Broadcast News Transcription and Understanding Workshop (pp 41-48)
- CU-HTK Acoustic modeling experiments. Hub5 Workshop 98
- Segment Generation and Clustering in the HTK Broadcast News Transcription System. 1998 DARPA Broadcast News Transcription and Understanding Workshop (pp 133-137)
- Experiments in broadcast news transcription. PROCEEDINGS OF THE 1998 IEEE INTERNATIONAL CONFERENCE ON ACOUSTICS, SPEECH AND SIGNAL PROCESSING, VOLS 1-6 (pp 909-912)
- On the convergence of fractal transforms. ICASSP’94 (pp 561-564)
- Utterance Weighted Multi-Dilation Temporal Convolutional Networks for Monaural Speech Dereverberation. Utterance Weighted Multi-Dilation Temporal Convolutional Networks for Monaural Speech Dereverberation, 1 September 2022. View this article in WRRO
- Receptive Field Analysis of Temporal Convolutional
Networks for Monaural Speech Dereverberation. IEEE 30th European Signal Processing Conference
- Non-Linear Pairwise Language Mappings for Low-Resource Multilingual Acoustic Model Fusion. Interspeech 2022, 23rd Annual Conference of the International Speech Communication Association
- Investigating the Impact of Cross-lingual Acoustic-Phonetic Similarities on Multilingual Speech Recognition. Interspeech 2022, 23rd Annual Conference of the International Speech Communication Association
- Insights on Neural Representations for End-to-End Speech Recognition. Interspeech 2021
- Empirical Interpretation of Speech Emotion Perception with Attention Based Model for Speech Emotion Recognition. Interspeech 2020
- Uncertainty-Aware Machine Support for Paper Reviewing on the Interspeech 2019 Submission Corpus. Interspeech 2020
- Unsupervised Acoustic Unit Representation Learning for Voice Conversion Using WaveNet Auto-Encoders. Interspeech 2020
- Multilingual Speech Recognition Using Language-Specific Phoneme Recognition as Auxiliary Task for Indian Languages. Interspeech 2020
- Detecting Mismatch Between Speech and Transcription Using Cross-Modal Attention. Interspeech 2019
- Latent Dirichlet Allocation Based Acoustic Data Selection for Automatic Speech Recognition. Interspeech 2019
- Learning Temporal Clusters Using Capsule Routing for Speech Emotion Recognition. Interspeech 2019
- On the Usefulness of the Speech Phase Spectrum for Pitch Extraction. Interspeech 2018 View this article in WRRO
- Channel Compensation in the Generalised Vector Taylor Series Approach to Robust ASR. Interspeech 2017 View this article in WRRO
- Combining Feature and Model-Based Adaptation of RNNLMs for Multi-Genre Broadcast Speech Recognition. Interspeech 2016 View this article in WRRO
- Colloquialising Modern Standard Arabic Text for Improved Speech Recognition. Interspeech 2016 View this article in WRRO
- The Sheffield Wargame Corpus — Day Two and Day Three. Interspeech 2016 View this article in WRRO
- Improving Generalisation to New Speakers in Spoken Dialogue State Tracking. Interspeech 2016 View this article in WRRO
- DNN-Based Speaker Clustering for Speaker Diarisation. Interspeech 2016 View this article in WRRO
- Automatic Genre and Show Identification of Broadcast Media. Interspeech 2016 View this article in WRRO
- A Novel Phase-based Feature for Robust Speech Recognition. USES Conference Proceedings
Reports
- Clarity Prediction Challenge 1 Entry: Non-intrusive Speech Intelligibility Metric Prediction - Technical Report
- Data Dependence of Speech Decoder Parameters
- Confidence-informed unsupervised Minimum Bayes Risk acoustic model adaptation
- Conversational multi-party speech recognition using remote microphones
- Automatic Transcription of Conversational Telephone Speech. Development of the CU-HTK 2002 System
Theses / Dissertations
- Hidden Model Sequence Models for Automatic Speech Recognition.
- On the Use of Iterated Function Systems for Coding of Grayscale Images.
Datasets
- The homeService corpus v1.0.
- Experiments results for IEEE/ACM Transaction on Audio, Speech and Language Processing Journal Paper: "Recurrent Neural Network Language Model Adaptation for Multi-Genre Broadcast Speech Recognition and Alignment".
- Interspeech 2016 - Experiment results for paper "Error correction in lightly supervised alignment of broadcast subtitles".
- Interspeech 2016 - Experiment results for paper "webASR 2 - Improved cloud based speech technology".
- Computer, Speech and Language - Experiment results for paper "Acoustic Adaptation to Dynamic Background Conditions with Asynchronous Transformations".
- Multimedia Tools and Applications - Experiments results for paper "Lightly supervised alignment of subtitles on multigenre broadcasts".
- Interspeech 2016 - Experiment results for paper "Combining Feature and Model-Based Adaptation of RNNLMs for Multi-Genre Broadcast Speech Recognition".
- Interspeech 2016 - Experiment results for Sheffield Wargame Corpora (SWC1, SWC2, SWC3).
- ICASSP 2016 - Experiment results for the paper "Groupwise learning for ASR k-best list reranking in spoken language translation".
Other
Preprints
- Towards domain generalisation in ASR with elitist sampling and ensemble knowledge distillation, arXiv.
- Perceive and predict: self-supervised speech representation based loss
functions for speech enhancement.
- Dynamic Kernels and Channel Attention with Multi-Layer Embedding Aggregation for Speaker Verification, arXiv.
- Probing Statistical Representations For End-To-End ASR, arXiv.
- Deformable Temporal Convolutional Networks for Monaural Noisy
Reverberant Speech Separation.
- Unsupervised data selection for Speech Recognition with contrastive loss
ratios.
- Non-Linear Pairwise Language Mappings for Low-Resource Multilingual
Acoustic Model Fusion.
- Investigating the Impact of Cross-lingual Acoustic-Phonetic Similarities
on Multilingual Speech Recognition.
- A cross-corpus study on speech emotion recognition.
- Insights on Neural Representations for End-to-End Speech Recognition, arXiv.
- Utterance Weighted Multi-Dilation Temporal Convolutional Networks for
Monaural Speech Dereverberation.
- Receptive Field Analysis of Temporal Convolutional Networks for Monaural
Speech Dereverberation.
- Efficient Non-Autoregressive GAN Voice Conversion using VQWav2vec
Features and Dynamic Convolution.
- MetricGAN+/-: Increasing Robustness of Noise Reduction on Unseen Data. View this article in WRRO
- Multiple-hypothesis CTC-based semi-supervised adaptation of end-to-end
speech recognition, arXiv.
- Towards Low-Resource StarGAN Voice Conversion using Weight Adaptive
Instance Normalization.
- Unsupervised Acoustic Unit Representation Learning for Voice Conversion
using WaveNet Auto-encoders.
- Automatic Genre and Show Identification of Broadcast Media, arXiv.
- The 2015 Sheffield System for Transcription of Multi-Genre Broadcast
Media, arXiv.
- Latent Dirichlet Allocation Based Organisation of Broadcast Media
Archives for Deep Neural Network Adaptation, arXiv.
- Background-tracking Acoustic Features for Genre Identification of
Broadcast Shows, arXiv.
- Data-selective Transfer Learning for Multi-Domain Speech Recognition, arXiv.
- Unsupervised Domain Discovery using Latent Dirichlet Allocation for
Acoustic Modelling in Speech Recognition, arXiv.
- The USFD Spoken Language Translation System for IWSLT 2014. View this article in WRRO
- The HTK Book. Cambridge, England: Cambridge University Engineering Department.
- Grants
-
Current grants
- Automatic voice conversion for transforming professional adult voice actors to artificial child voice actors, Innovate UK, 01/2021 - 01/2023, £173,605, as PI
- UKRI Centre for Doctoral Training in Speech and Language Technologies and their Applications, EPSRC, 04/2019 to 09/2027, £5,508,850, as PI
- VoiceBase Centre, VoiceBase Inc., 04/2018 - 03/2022, £1,499,972, as PI
- WFST-based integration of ASR and MT in Spoken Language Translation, Google, 03/2014 to 12/2022, £63,588, as PI
Previous grants
- MAUDIE: Multimedia Analysis for Unsupervised Dubbing In Entertainment, Innovate UK, 05/2018 to 07/2021, £393,115, as PI
- TUTO II: Reading skills tutoring system, ITSLANGUAGE BV, 08/2017 to 12/2019, £121,439, as PI
- Sound Source Separation Based on Deep Learning, Industrial, 05/2019 - 04/2020, £48,000, as PI
- Acoustic correlates of emotions for automatic recognition, Industrial, 10/2018 to 09/2019, £48,900, as PI
- Bridge Project, VoiceBase Inc., 09/2017 to 03/2018, £61,200, as PI
- STATUS IV: Speech Technology and Translation Universal Survey, Defence Science and Technology Laboratory, 01/2017 to 10/2017, £60,000, as PI
- TUTO: Reading skills tutoring system, ITSLANGUAGE BV, 09/2016 to 08/2017, £61,983, as PI
- STATUS III: Speech Technology and Translation Universal Survey, Defence Science and Technology Laboratory, 01/2015 to 07/2016, £78,684, as PI
- STATUS II: Speech Technology and Translation Universal Survey, Defence Science and Technology Laboratory, 11/2013 to 05/2014, £98,982, as PI
- ItsLanguage, ITSLANGUAGE BV, 11/2012 to 03/2015, £68,333, as PI
- German System Adaptation, ITSLANGUAGE BV, 11/2012 to 03/2015, £42,373, as PI
- DocuMeet: Transcription, summarisation and documentation of meetings using advanced speech technologies, indexing and browsing capabilities, European Commission - FP7, 11/2012 to 10/2014, £368,433, as PI
- STATUS: Speech Technology and Translation Universal Survey, Defence Science and Technology Laboratory, 10/2012 to 08/2013, £73,726, as PI
- A Joint Model of Spoken Language Translation, Google, 09/2011 to 12/2016, £43,014, as PI
- Natural Speech Technology, EPSRC, 05/2011 to 07/2016, £1,798,665, as PI
- Unsupervised Domain Adaptation, CISCO, 11/2010 to 04/2012, £121,745, as PI
- AMIDA: Augmented Multi-party Interaction with Distance Access, European Commission - FP6, 10/2006 to 12/2009, £467,074, as PI
- AMIDA: Augmented Multi-party Interaction with Distance Access, European Commission - FP6, 10/2006 to 12/2009, £345,350, as PI
- Professional activities and memberships
-
- Head of the Speech and Hearing research group
- Editorial Board member, Computer Speech and Language
- Associate Editor, ACM Transactions on Speech and Language Processing
- Organising committee member, ASRU 2013
- Area Chair, Interspeech 2014, Speech Recognition - Signal Processing, Acoustic Modelling, Robustness and Adaptation.
- Area Chair, ICPR 2014, Track 3 Image, Speech. Signal and Video Processing
- Programme Committee, PoITAL 2014