• Publications
  • Influence
Merlin: An Open Source Neural Network Speech Synthesis System
TLDR
We introduce the Merlin speech synthesis toolkit for neural network-based speech synthesis. Expand
  • 242
  • 34
  • PDF
Detection of phonological features in continuous speech using neural networks
TLDR
We report work on the first component of a two-stage speech recognition architecture based onphonological features rather than phones. Expand
  • 200
  • 22
  • PDF
Over-exposed?: privacy patterns and considerations in online and mobile photo sharing
TLDR
We use context-aware camerephone devices to examine privacy decisions in mobile and online photo sharing and identify relationships between location of photo capture and privacy settings. Expand
  • 321
  • 16
  • PDF
The fouling of microfiltration membranes by NOM after coagulation treatment
Microfiltration membranes used in drinking-water treatment are fouled by both colloidal material and natural organic matter (NOM) present in the raw water. The relative importance of theseExpand
  • 369
  • 15
  • PDF
Multisyn: Open-domain unit selection for the Festival speech synthesis system
TLDR
We present the implementation and evaluation of an open-domain unit selection speech synthesis engine designed to be flexible enough to encourage further unit selection research and allow rapid voice development by users with minimal speech synthesis knowledge. Expand
  • 155
  • 15
  • PDF
Speech production knowledge in automatic speech recognition.
TLDR
A survey of a growing body of work in which such representations are used to improve automatic speech recognition is provided. Expand
  • 206
  • 13
  • PDF
Announcing the Electromagnetic Articulography (Day 1) Subset of the mngu0 Articulatory Corpus
TLDR
This paper serves as an initial announcement of the availability of a corpus of articulatory data called mngu0, which we hope will be useful to other researchers and will come to be equally widely-used. Expand
  • 98
  • 13
  • PDF
A study of speaker adaptation for DNN-based speech synthesis
TLDR
A major advantage of statistical parametric speech synthesis over unit-selection speech synthesis is its adaptability and controllability in changing speaker characteristics and speaking style. Expand
  • 105
  • 11
  • PDF
Modelling the uncertainty in recovering articulation from acoustics
TLDR
This paper presents an experimental comparison of the performance of the multilayer perceptron (MLP) with that of the mixture density network (MDN) for an acoustic-to-articulatory mapping task. Expand
  • 98
  • 11
  • PDF
Articulatory feature recognition using dynamic Bayesian networks
TLDR
We describe a dynamic Bayesian network for articulatory feature recognition. Expand
  • 111
  • 10
  • PDF