Continuous speech recognition for the TIMIT database using neural networks

Four types of neural networks which have previously been established for speech recognition and tested on a small, seven-speaker, 100-sentence database are applied to the TIMIT database. The networks are a recurrent network phoneme recognizer, a modified Kanerva model morph recognizer, a compositional representation phoneme-to-word recognizer, and a modified Kanerva model morph-to-word recognizer. The major result is for the recurrent net, giving a phoneme recognition accuracy of 57% from the si and sx sentences. The Kanerva morph recognizer achieves 66.2% accuracy for a small subset of the sa and sx sentences. The results for the word recognizers are incomplete.

MMI training for continuous phoneme recognition on the TIMIT database

This paper reports our experiences with a phoneme recognition system for the TIMIT database which uses multiple mixture continuous density monophone HMMs trained using MMI. A comprehensive set of results are presented comparing the ML and MMI training criteria for both diagonal and full covariance models. These results using simple monophone HMMs show clear performance gains achieved by MMI training, and are comparable to the best reported by others including those which use context-dependent models. In addition, the paper discusses a number of performance and implementation issues which are crucial to successful MMI training.

Continuous F0 in the source-excitation generation for HMM-based TTS: Do we need voiced/unvoiced classification?

Most HMM-based TTS systems use a hard voiced/unvoiced classification to produce a discontinuous F0 signal which is used for the generation of the source-excitation. When a mixed source excitation is used, this decision can be based on two different sources of information: the state-specific MSD-prior of the F0 models, and/or the frame-specific features generated by the aperiodicity model. This paper examines the meaning of these variables in the synthesis process, their interaction, and how they affect the perceived quality of the generated speech The results of several perceptual experiments show that when using mixed excitation, subjects consistently prefer samples with very few or no false unvoiced errors, whereas a reduction in the rate of false voiced errors does not produce any perceptual improvement. This suggests that rather than using any form of hard voiced/unvoiced classification, e.g., the MSD-prior, it is better for synthesis to use a continuous F0 signal and rely on the frame-level soft voiced/unvoiced decision of the aperiodicity model. © 2011 IEEE.

Joint modelling of voicing label and continuous F0 for HMM based speech synthesis

Fundamental frequency, or F0 is critical for high quality speech synthesis in HMM based speech synthesis. Traditionally, F0 values are considered to depend on a binary voicing decision such that they are continuous in voiced regions and undefined in unvoiced regions. Multi-space distribution HMM (MSDHMM) has been used for modelling the discontinuous F0. Recently, a continuous F0 modelling framework has been proposed and shown to be effective, where continuous F0 observations are assumed to always exist and voicing labels are explicitly modelled by an independent stream. In this paper, a refined continuous F0 modelling approach is proposed. Here, F0 values are assumed to be dependent on voicing labels and both are jointly modelled in a single stream. Due to the enforced dependency, the new method can effectively reduce the voicing classification error. Subjective listening tests also demonstrate that the new approach can yield significant improvements on the naturalness of the synthesised speech. A dynamic random unvoiced F0 generation method is also investigated. Experiments show that it has significant effect on the quality of synthesised speech. © 2011 IEEE.

Structured discriminative models for noise robust continuous speech recognition

Recently there has been interest in structured discriminative models for speech recognition. In these models sentence posteriors are directly modelled, given a set of features extracted from the observation sequence, and hypothesised word sequence. In previous work these discriminative models have been combined with features derived from generative models for noise-robust speech recognition for continuous digits. This paper extends this work to medium to large vocabulary tasks. The form of the score-space extracted using the generative models, and parameter tying of the discriminative model, are both discussed. Update formulae for both conditional maximum likelihood and minimum Bayes' risk training are described. Experimental results are presented on small and medium to large vocabulary noise-corrupted speech recognition tasks: AURORA 2 and 4. © 2011 IEEE.

Simulated Annealing: Rigorous finite-time guarantees for optimization on continuous domains

Simulated annealing is a popular method for approaching the solution of a global optimization problem. Existing results on its performance apply to discrete combinatorial optimization where the optimization variables can assume only a finite set of possible values. We introduce a new general formulation of simulated annealing which allows one to guarantee finite-time performance in the optimization of functions of continuous variables. The results hold universally for any optimization problem on a bounded domain and establish a connection between simulated annealing and up-to-date theory of convergence of Markov chain Monte Carlo methods on continuous domains. This work is inspired by the concept of finite-time learning with known accuracy and confidence developed in statistical learning theory.

Algorithm-based continuous pulse duration tuning and performance control of a mode-locked laser diode.

A control algorithm is presented that addresses the stability issues inherent to the operation of monolithic mode-locked laser diodes. It enables a continuous pulse duration tuning without any onset of Q-switching instabilities. A demonstration of the algorithm performance is presented for two radically different laser diode geometries and continuous pulse duration tuning between 0.5 ps to 2.2 ps and 1.2 ps to 10.2 ps is achieved. With practical applications in mind, this algorithm also facilitates control over performance parameters such as output power and wavelength during pulse duration tuning. The developed algorithm enables the user to harness the operational flexibility from such a laser with 'push-button' simplicity.

Reinforcement learning with reference tracking control in continuous state spaces

The contribution described in this paper is an algorithm for learning nonlinear, reference tracking, control policies given no prior knowledge of the dynamical system and limited interaction with the system through the learning process. Concepts from the field of reinforcement learning, Bayesian statistics and classical control have been brought together in the formulation of this algorithm which can be viewed as a form of indirect self tuning regulator. On the task of reference tracking using a simulated inverted pendulum it was shown to yield generally improved performance on the best controller derived from the standard linear quadratic method using only 30 s of total interaction with the system. Finally, the algorithm was shown to work on the simulated double pendulum proving its ability to solve nontrivial control tasks. © 2011 IEEE.

Partially Observable Markov Decision Processes with continuous observations for dialogue management

This work shows how a dialogue model can be represented as a Partially Observable Markov Decision Process (POMDP) with observations composed of a discrete and continuous component. The continuous component enables the model to directly incorporate a confidence score for automated planning. Using a testbed simulated dialogue management problem, we show how recent optimization techniques are able to find a policy for this continuous POMDP which outperforms a traditional MDP approach. Further, we present a method for automatically improving handcrafted dialogue managers by incorporating POMDP belief state monitoring, including confidence score information. Experiments on the testbed system show significant improvements for several example handcrafted dialogue managers across a range of operating conditions.

Continuous Diffraction Patterns from Circular Arrays of Carbon Nanotubes

We report the remarkable diffraction effects produced from circular patterned arrays of multiwalled carbon nanotubes (MWCNTs). Highly ordered circular arrays of multiwalled carbon nanotubes (with inter-nanotube spacings of 633 nm) display optical dispersion effects similar to compact discs. These arrays display remarkable diffraction patterns in the far field which are spatially continuous. High quality diffraction patterns were obtained experimentally which are in excellent agreement with the theoretical calculations. The achieved continuous diffraction patterns pave the way towards the utilization of engineered carbon nanotube arrays in applications like three dimensional holograms.

On the transition from finite-volume negatively buoyant releases to continuous fountains

An experimental investigation to identify the source conditions that distinguish finite-volume negatively buoyant fluid projectile behaviour from fountain behaviour in quiescent environments of uniform density is described. Finite-volume releases are governed by their source Froude number Fr D and the aspect ratio L/D of the release, where L denotes the length of the column of fluid dispensed vertically from the nozzle of diameter D. We establish the influence of L/D on the peak rise heights of a release formed by dispensing saline solution into fresh water for 0continuous flux. The critical aspect ratio (L/D) f, for a given Fr D, which when exceeded no longer influenced release behaviour, led to the determination of Fr D, (L/D) f paired source conditions that give rise to solely Froude-number-dependent, i.e. fountain-like, behaviour. As such, we make the link between finite-volume releases and continuous fountains. The Fr D(L/D) f pairs led us directly to the classification of a Fr D, L/D space from which source conditions giving rise to either negatively buoyant projectiles or fountains may be readily identified. The variation of (L/D) f with Fr D corresponds closely to established fountain regimes of very weak, weak and forced fountains. Moreover, our results indicate that the formation or otherwise of a primary vortex, as fluid is ejected, has a profound influence on the length of the dispensed fluid column that is necessary to achieve rise heights equal to fountain rise heights. © 2012 Cambridge University Press.