Datasets

• CSTR VCTK Corpus: English Multi-speaker Corpus for CSTR Voice Cloning Toolkit (version 0.92)

  Dataset

• Experiment materials for "The temporal delay hypothesis: Natural, vocoded and synthetic speech."

  Dataset

• Artificial Personality

  Dataset

• Human vs Machine Spoofing

  Dataset

• Experiment materials for "Disfluencies in change detection in natural, vocoded and synthetic speech."

  Dataset

Activities

• Controllable and adaptable statistical parametric speech synthesis

  Activity: Academic talk or presentation types › Invited talk

• ICPhS 2015 keynote: What speech synthesis can do for you (and what you can do for speech synthesis)

  Activity: Academic talk or presentation types › Invited talk

• Neural Networks for Distant Speech Recognition

  Activity: Academic talk or presentation types › Invited talk

• Learning representations of speech in neural network acoustic models

  Activity: Academic talk or presentation types › Invited talk

• EACL 2014 keynote: Speech synthesis needs YOU!

  Activity: Academic talk or presentation types › Invited talk

• Sunday Morning With...

  Activity: Other activity types › Types of Public engagement and outreach - Media article or participation

Press/Media

• Interview with Professor Steve Renals on project to improve talking gadgets

  Press/Media: Research

• Prof Steve Renals leads project to improve talking gadgets

  Press/Media: Research

• Prof Simon King and Dr Philipa Rewaj discuss the Speak:Unique project

  Press/Media: Public Engagement Activities

Research outputs

• Reconstructing Voices within the Multiple-Average-Voice-Model framework

  Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

• Multi-frame factorisation for long-span acoustic modelling

  Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

• Human vs Machine Spoofing Detection on Wideband and Narrowband Data

  Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

• Determining the number of speakers in a meeting using microphone array features

  Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

• On the effect of SNR and superdirective beamforming in speaker diarisation in meetings

  Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

• Improving Lightly Supervised Training for Broadcast Transcriptions

  Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

• The Sheffield Wargames Corpus.

  Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

• Fast, low-artifact speech synthesis considering global variance

  Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

• Noise adaptive training for subspace Gaussian mixture models

  Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

• Noise Compensation for Subspace Gaussian Mixture Models.

  Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

• A Comparative Study of Adaptive, Automatic recognition of Disordered Speech

  Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

• Transcription of multi-genre media archives using out-of-domain data

  Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

• Description of the UEDIN System for German ASR

  Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

• The UEDIN ASR Systems for the IWSLT 2014 Evaluation

  Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

• Towards Personalized Synthesized Voices for Individuals with Vocal Disabilities: Voice Banking and Reconstruction

  Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

• A Lattice-based Approach to Automatic Filled Pause Insertion

  Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

• The Temporal Delay Hypothesis: Natural, Vocoded and Synthetic Speech

  Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

• Combining a Vector Space Representation of Linguistic Context with a Deep Neural Network for Text-To-Speech Synthesis

  Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

• Learning Hidden Unit Contributions for Unsupervised Speaker Adaptation of Neural Network Acoustic Models

  Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

• Speech Synthesis

  Research output: Chapter in Book/Report/Conference proceeding › Chapter

• Investigating Automatic & Human Filled Pause Insertion for Speech Synthesis

  Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

• Measuring a decade of progress in Text-to-Speech

  Research output: Contribution to journal › Article › peer-review

• Investigating the shortcomings of HMM synthesis

  Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

• Joint Uncertainty Decoding with Unscented Transform for Noise Robust Subspace Gaussian Mixture Models

  Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

• homeService: Voice-enabled assistive technology in the home using cloud-based automatic speech recognition

  Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

• The NST–GlottHMM entry to the Blizzard Challenge 2015

  Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

• Waveform generation based on signal reshaping for statistical parametric speech synthesis

  Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

• Synthesising Filled Pauses: Representation and Datamixing

  Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

• Median-based generation of synthetic speech durations using a non-parametric approach

  Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

• Using eigenvoices and nearest-neighbours in HMM-based cross-lingual speaker adaptation with limited data

  Research output: Contribution to journal › Article › peer-review

• Investigation of Using Continuous Representation of Various Linguistic Units in Neural Network Based Text-to-Speech Synthesis

  Research output: Contribution to journal › Article › peer-review

• Enhance the Word Vector with Prosodic Information for the Recurrent Neural Network Based TTS System

  Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

• A Hierarchical Predictor of Synthetic Speech Naturalness Using Neural Networks

  Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

• Analysis of the Voice Conversion Challenge 2016 Evaluation Results

  Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

• Development of a statistical parametric synthesis system for operatic singing in German

  Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

• Multidimensional scaling of systems in the Voice Conversion Challenge 2016

  Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

• Investigating RNN-based speech enhancement methods for noise-robust Text-to-Speech

  Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

• Learning Hidden Unit Contributions for Unsupervised Acoustic Model Adaptation

  Research output: Contribution to journal › Article › peer-review

• A deep auto-encoder based low-dimensional feature extraction from FFT spectral envelopes for statistical parametric speech synthesis

  Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

• Hybrid acoustic models for distant and multichannel large vocabulary speech recognition

  Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

• A perceptually-motivated low-complexity instantaneous linear channel normalization technique applied to speaker verification

  Research output: Contribution to journal › Article › peer-review

• Voice Banking and Voice Reconstruction for MND patients

  Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

• Cross-Lingual Subspace Gaussian Mixture Models for Low-Resource Speech Recognition

  Research output: Contribution to journal › Article › peer-review

• Smooth talking: articulatory join costs for unit selection

  Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

• Evaluating comprehension of natural and synthetic conversational speech

  Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

• Simplified Learning Using Binary Orthogonal Constraints

  Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

• Cross-Lingual Adaptation with Multi-Task Adaptive Networks

  Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

• Speaker Recognition Anti-spoofing

  Research output: Chapter in Book/Report/Conference proceeding › Chapter

• Towards minimum perceptual error training for DNN-based speech synthesis

  Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

• Voice source modelling using deep neural networks for statistical parametric speech synthesis

  Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

• Soft context clustering for F0 modeling in HMM-based speech synthesis

  Research output: Contribution to journal › Article › peer-review

• Asynchronous factorisation of speaker and background with feature transforms in speech recognition

  Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

• Lightly Supervised Discriminative Training of Grapheme Models for Improved Sentence-level Alignment of Speech and Text Data

  Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

• Automatic Transcription of Multi-genre Media Archives

  Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

• The UEDIN English ASR System for the IWSLT 2013 Evaluation

  Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

• The CSTR entry to the Blizzard Challenge 2016

  Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

• Probabilistic Linear Discriminant Analysis with Bottleneck Features for Speech Recognition

  Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

• Differentiable pooling for unsupervised speaker adaptation

  Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

• A Flexible Front-End for HTS

  Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

• Feature-space Speaker Adaptation for Probabilistic Linear Discriminant Analysis Acoustic Models

  Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

• Modelling acoustic feature dependencies with artificial neural networks: Trajectory-RNADE

  Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

• Dysarthria Intelligibility Assessment in a Factor Analysis Total Variability Space

  Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

• A lecture transcription system combining neural network acoustic and language models

  Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

• Revisiting semi-continuous hidden Markov models

  Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

• Using Bayesian Networks to find relevant context features for HMM-based speech synthesis

  Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

• Factorized context modelling for Text-to-Speech synthesis

  Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

• Evaluation of Speaker Verification Security and Detection of HMM-Based Synthetic Speech

  Research output: Contribution to journal › Article › peer-review

• Multitask Learning of Context-Dependent Targets in Deep Neural Network Acoustic Models

  Research output: Contribution to journal › Article › peer-review

• Deep neural networks employing multi-task learning and stacked bottleneck features for speech synthesis.

  Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

• Anti-Spoofing for Text-Independent Speaker Verification: An Initial Database, Comparison of Countermeasures, and Human Performance

  Research output: Contribution to journal › Article › peer-review

• A Deep Generative Architecture for Postfiltering in Statistical Parametric Speech Synthesis

  Research output: Contribution to journal › Article › peer-review

• The MGB Challenge: Evaluating Multi-Genre Broadcast Media Recognition

  Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

• Sentence-level control vectors for deep neural network speech synthesis

  Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

• Are we using enough listeners? No! An empirically-supported critique of Interspeech 2014 TTS evaluations

  Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

• Maximum a posteriori adaptation of subspace Gaussian mixture models for cross-lingual speech recognition

  Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

• Multilingual training of deep neural networks

  Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

• Speech and Audio Signal Processing: Processing and Perception of Speech and Music

  Research output: Book/Report › Book

• Neural net word representations for phrase-break prediction without a part of speech tagger

  Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

• Multiple-average-voice-based speech synthesis

  Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

• Unsupervised Adaptation of Recurrent Neural Network Language Models

  Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

• The NII speech synthesis entry for Blizzard Challenge 2016

  Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

• Small-footprint Deep Neural Networks with Highway Connections for Speech Recognition

  Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

• Intelligibility enhancement of HMM-generated speech in additive noise by modifying Mel cepstral coefficients to increase the glimpse proportion

  Research output: Contribution to journal › Article › peer-review

• Deep neural network-guided unit selection synthesis

  Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

• Using linguistic predictability and the Lombard effect to increase the intelligibility of synthetic speech in noise

  Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

• Lightly supervised learning from a damaged natural speech corpus

  Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

• Multi-level adaptive networks in tandem and hybrid ASR systems

  Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

• Analysis of Speaker Clustering Strategies for HMM-Based Speech Synthesis

  Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

• The UEDIN system for the IWSLT 2012 evaluation

  Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

• Intelligibility Enhancement of Speech in Noise

  Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

• A reading list of recent advances in speech synthesis

  Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

• Disfluencies in change detection in natural, vocoded and synthetic speech

  Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

• Speech Synthesis

  Research output: Chapter in Book/Report/Conference proceeding › Chapter (peer-reviewed)

• Studio report: Linux audio for multi-speaker natural speech technology.

  Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

• End-to-end neural segmental models for speech recognition

  Research output: Contribution to journal › Article › peer-review

• Parallel and cascaded deep neural networks for text-to-speech synthesis

  Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

• Syllable-Level Representations of Suprasegmental Features for DNN-Based Text-to-Speech Synthesis

  Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

• GlottDNN - A full-band glottal vocoder for statistical parametric speech synthesis

  Research output: Contribution to conference › Paper › peer-review

• Neural networks for distant speech recognition

  Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

• ASVspoof: the Automatic Speaker Verification Spoofing and Countermeasures Challenge

  Research output: Contribution to journal › Article › peer-review

• A Comparative Study of the Performance of HMM, DNN, and RNN based Speech Synthesis Systems Trained on Very Large Speaker-Dependent Corpora

  Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

• Robust TTS Duration Modelling Using DNNs

  Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

• From HMMs to DNNs: Where Do the Improvements Come From?

  Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

• Sequence-discriminative training of deep neural networks

  Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

• Unsupervised cross-lingual knowledge transfer in DNN-based LVCSR

  Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

• Generating exact lattices in the WFST framework

  Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

• Cross-domain Paraphrasing For Improving Language Modelling Using Out-of-domain Data

  Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

• Merlin: An Open Source Neural Network Speech Synthesis System

  Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

• Investigating Very Deep Highway Networks for Parametric Speech Synthesis

  Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

• Differentiable Pooling for Unsupervised Acoustic Model Adaptation

  Research output: Contribution to journal › Article › peer-review

• The Effect of Filled Pauses and Speaking Rate on Speech Comprehension in Natural, Vocoded and Synthetic Speech

  Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

• A system for automatic broadcast news summarisation, geolocation and translation

  Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

• Deep neural network context embeddings for model selection in rich-context HMM synthesis

  Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

• Convolutional Neural Networks for Distant Speech Recognition

  Research output: Contribution to journal › Article › peer-review

• Grapheme and multilingual posterior features for under-resourced speech recognition: a study on Scottish Gaelic

  Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

• Speech Enhancement for a Noise-Robust Text-to-Speech Synthesis System using Deep Recurrent Neural Networks

  Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

• Complementary tasks for context-dependent deep neural network acoustic models

  Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

• A study of speaker adaptation for DNN-based speech synthesis

  Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

• Distant Speech Recognition Experiments Using the AMI Corpus

  Research output: Chapter in Book/Report/Conference proceeding › Chapter (peer-reviewed) › peer-review

• A system for automatic alignment of broadcast media captions using weighted finite-state transducers

  Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

• SAT-LHUC: Speaker adaptive training for learning hidden unit contributions

  Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

• Combining Vocal Tract Length Normalization With Hierarchical Linear Transformations

  Research output: Contribution to journal › Article › peer-review

• SAS: A Speaker Verification Spoofing Database Containing Diverse Attacks

  Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

• The Voice Conversion Challenge 2016

  Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

• Measuring the Perceptual Effects of Modelling Assumptions in Speech Synthesis Using Stimuli Constructed from Repeated Natural Speech

  Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

• Regularization of context-dependent deep neural networks with context-independent multi-task training

  Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

• Acoustic Data-driven Pronunciation Lexicon for Large Vocabulary Speech Recognition

  Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

• Blind Speech Segmentation using Spectrogram Image-based Features and Mel Cepstral Coefficients

  Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

• The use of articulatory movement data in speech synthesis applications: An overview — Application of articulatory movements using machine learning algorithms

  Research output: Contribution to journal › Article › peer-review

• Artificial Personality and Disfluency

  Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

• Knowledge Distillation for Small-footprint Highway Networks

  Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

• Using Text and Acoustic Features in Predicting Glottal Excitation Waveforms for Parametric Speech Synthesis with Recurrent Neural Networks

  Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

• Just-in-time prepared captioning for live transmissions

  Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

• Segmental Recurrent Neural Networks for End-to-end Speech Recognition

  Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

• A Study of the Recurrent Neural Network Encoder-Decoder for Large Vocabulary Speech Recognition

  Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

• Rating Naturalness in Speech Synthesis: The Effect of Style and Expectation

  Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

• Probabilistic Linear Discriminant Analysis for Acoustic Modeling

  Research output: Contribution to journal › Article › peer-review

• SPECS - an embedded platform, speech-driven environmental control system evaluated in a virtuous circle framework

  Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

• Speech Enhancement of Noisy and Reverberant Speech for Text-to-Speech

  Research output: Contribution to journal › Article › peer-review

• An introduction to statistical parametric speech synthesis

  Research output: Contribution to journal › Article › peer-review

• Joint Uncertainty Decoding for Noise Robust Subspace Gaussian Mixture Models

  Research output: Contribution to journal › Article › peer-review

• Speech Synthesis Based on Hidden Markov Models

  Research output: Contribution to journal › Article › peer-review

• Glottal Spectral Separation for Speech Synthesis

  Research output: Contribution to journal › Article › peer-review

• Small-footprint highway deep neural networks for speech recognition

  Research output: Contribution to journal › Article › peer-review

• Top-down Tree Long Short-Term Memory Networks

  Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

• Investigating gated recurrent neural networks for speech synthesis

  Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

• On training the recurrent neural network encoder-decoder for large vocabulary end-to-end speech recognition

  Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

• Structured Output Layer with Auxiliary Targets for Context-Dependent Acoustic Modelling

  Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

• ASVspoof 2015: the First Automatic Speaker Verification Spoofing and Countermeasures Challenge

  Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

• Combining in-domain and out-of-domain speech data for automatic recognition of disordered speech

  Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

• Learning speaker-specific pronunciations of disordered speech

  Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

• Testing the Consistency Assumption: Pronunciation Variant Forced Alignment in Read and Spontaneous Speech Synthesis

  Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

• Speech synthesis technologies for individuals with vocal disabilities: Voice banking and reconstruction

  Research output: Contribution to journal › Article › peer-review

• Autoregressive Models for Statistical Parametric Speech Synthesis

  Research output: Contribution to journal › Article › peer-review

• Regularized subspace Gaussian mixture models for cross-lingual speech recognition

  Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

• Revisiting Hybrid and GMM-HMM system combination techniques

  Research output: Chapter in Book/Report/Conference proceeding › Conference contribution