Publications by Thomas Merritt

Deep neural networks (DNNs) have recently been the focus of much text-to-speech research as a replacement for decision trees and hidden Markov models (HMMs) in statistical parametric synthesis systems. Performance improvements have been reported; however, the configuration of systems evaluated makes it impossible to judge how much of the improvement is due to the new machine learning methods, and how much is due to other novel aspects of the systems. Specifically, whereas the decision trees in HMM-based systems typically operate at the state-level, and separate trees are used to handle separate acoustic streams, most DNN-based systems are trained to make predictions simultaneously for all streams at the level of the acoustic frame. This paper isolates the influence of three factors (machine learning method; state vs. frame predictions; separate vs. combined stream predictions) by building a continuum of systems along which only a single factor is varied at a time. We find that replacing decision trees with DNNs and moving from state-level to frame-level predictions both significantly improve listeners' naturalness ratings of synthetic speech produced by the systems. No improvement is found to result from switching from separate-stream to combined-stream predictions.

Keywords: speech synthesis, hidden Markov model, decision tree, deep neural network

Vocoding of speech is a standard part of statistical parametric speech synthesis systems. It imposes an upper bound of the naturalness that can possibly be achieved. Hybrid systems using parametric models to guide the selection of natural speech units can combine the benefits of robust statistical models with the high level of naturalness of waveform concatenation. Existing hybrid systems use Hidden Markov Models (HMMs) as the statistical model. This paper demonstrates that the superiority of Deep Neural Network (DNN) acoustic models over HMMs in conventional statistical parametric speech synthesis also carries over to hybrid synthesis. We compare various DNN and HMM hybrid configurations, guiding the selection of waveform units in either the vocoder parameter domain, or in the domain of embeddings (bottleneck features).

This paper describes the text-to-speech system entered by The Centre for Speech Technology Research into the 2016 Blizzard Challenge. This system is a hybrid synthesis system which uses output from a recurrent neural network to drive a unit selection synthesiser. The annual Blizzard Challenge conducts side-by-side testing of a number of speech synthesis systems trained on a common set of speech data. The task of the 2016 Blizzard Challenge is to train on expressively-read children’s storybooks, and to synthesise speech in the same domain. The Challenge therefore presents an opportunity to test the effectiveness of several techniques we have developed when applied to expressive speech data.

This paper introduces a novel form of parametric synthesis that uses context embeddings produced by the bottleneck layer of a deep neural network to guide the selection of models in a rich-context HMM-based synthesiser. Rich-context synthesis – in which Gaussian distributions estimated from single linguistic contexts seen in the training data are used for synthesis, rather than more conventional decision tree-tied models – was originally proposed to address over-smoothing due to averaging across contexts. Our previous investigations have confirmed experimentally that averaging across different contexts is indeed one of the largest factors contributing to the limited quality of statistical parametric speech synthesis. However, a possible weakness of the rich context approach as previously formulated is that a conventional tied model is still used to guide selection of Gaussians at synthesis time. Our proposed approach replaces this with context embeddings derived from a neural network.

Even the best statistical parametric speech synthesis systems do not achieve the naturalness of good unit selection. We investigated possible causes of this. By constructing speech signals that lie inbetween natural speech and the output from a complete HMM synthesis system, we investigated various effects of modelling. We manipulated the temporal smoothness and the variance of the spectral parameters to create stimuli, then presented these to listeners alongside natural and vocoded speech, as well as output from a full HMM-based text-to-speech system and from an idealised `pseudo-HMM'. All speech signals, except the natural waveform, were created using vocoders employing one of two popular spectral parameterisations: Mel-Cepstra or Mel-Line Spectral Pairs. Listeners made `same or different' pairwise judgements, from which we generated a perceptual map using Multidimensional Scaling. We draw conclusions about which aspects of HMM synthesis are limiting the naturalness of the synthetic speech.

This paper presents an investigation of the separate perceptual degradations introduced by the modelling of source and filter features in statistical parametric speech synthesis. This is achieved using stimuli in which various permutations of natural, vocoded and modelled source and filter are combined, optionally with the addition of filter modifications (e.g. global variance or modulation spectrum scaling). We also examine the assumption of independence between source and filter parameters. Two complementary perceptual testing paradigms are adopted. In the first, we ask listeners to perform “same or different quality” judgements between pairs of stimuli from different configurations. In the second, we ask listeners to give an opinion score for individual stimuli. Combining the findings from these tests, we draw some conclusions regarding the relative contributions of source and filter to the currently rather limited naturalness of statistical parametric synthetic speech, and test whether current independence assumptions are justified.

Acoustic models used for statistical parametric speech synthesis typically incorporate many modelling assumptions. It is an open question to what extent these assumptions limit the naturalness of synthesised speech. To investigate this question, we recorded a speech corpus where each prompt was read aloud multiple times. By combining speech parameter trajectories extracted from different repetitions, we were able to quantify the perceptual effects of certain commonly used modelling assumptions. Subjective listening tests show that taking the source and filter parameters to be conditionally independent, or using diagonal covariance matrices, significantly limits the naturalness that can be achieved. Our experimental results also demonstrate the shortcomings of mean-based parameter generation.

Keywords: speech synthesis, acoustic modelling, stream independence, diagonal covariance matrices, repeated speech

Parametric speech synthesis techniques depend on full context acoustic models generated by language front-ends, which analyse linguistic and phonetic structure. HTS, the leading parametric synthesis system, can use a number of different front-ends to generate full context models for synthesis and training. In this paper we explore the use of a new text processing front-end that has been added to the speech recognition toolkit Kaldi as part of an ongoing project to produce a new parametric speech synthesis system, Idlak. The use of XML specification files, a modular design, and modern coding and testing approaches, make the Idlak front-end ideal for adding, altering and experimenting with the contexts used in full context acoustic models. The Idlak front-end was evaluated against the standard Festival front-end in the HTS system. Results from the Idlak front-end compare well with the more mature Festival front-end (Idlak - 2.83 MOS vs Festival - 2.85 MOS), although a slight reduction in naturalness perceived by non-native English speakers can be attributed to Festival’s insertion of non-punctuated pauses.

This paper presents the beginnings of a framework for formal testing of the causes of the current limited quality of HMM (Hidden Markov Model) speech synthesis. This framework separates each of the effects of modelling to observe their independent effects on vocoded speech parameters in order to address the issues that are restricting the progression to highly intelligible and natural-sounding speech synthesis. The simulated HMM synthesis conditions are performed on spectral speech parameters and tested via a pairwise listening test, asking listeners to perform a “same or different” judgement on the quality of the synthesised speech produced between these conditions. These responses are then processed using multidimensional scaling to identify the qualities in modelled speech that listeners are attending to and thus forms the basis of why they are distinguishable from natural speech. The future improvements to be made to the framework will finally be discussed which include the extension to more of the parameters modelled during speech synthesis.