Unsupervised intra-lingual and cross-lingual speaker adaptation for HMM-based speech synthesis using two-pass decision tree construction

Hidden Markov model (HMM)-based speech synthesis systems possess several advantages over concatenative synthesis systems. One such advantage is the relative ease with which HMM-based systems are adapted to speakers not present in the training dataset. Speaker adaptation methods used in the field of HMM-based automatic speech recognition (ASR) are adopted for this task. In the case of unsupervised speaker adaptation, previous work has used a supplementary set of acoustic models to estimate the transcription of the adaptation data. This paper first presents an approach to the unsupervised speaker adaptation task for HMM-based speech synthesis models which avoids the need for such supplementary acoustic models. This is achieved by defining a mapping between HMM-based synthesis models and ASR-style models, via a two-pass decision tree construction process. Second, it is shown that this mapping also enables unsupervised adaptation of HMM-based speech synthesis models without the need to perform linguistic analysis of the estimated transcription of the adaptation data. Third, this paper demonstrates how this technique lends itself to the task of unsupervised cross-lingual adaptation of HMM-based speech synthesis models, and explains the advantages of such an approach. Finally, listener evaluations reveal that the proposed unsupervised adaptation methods deliver performance approaching that of supervised adaptation.

The evolution of information while building cross-domain models of a design: A video experiment

Models capturing the connectivity between different domains of a design, e.g. between components and functions, can provide a tool for tracing and analysing aspects of that design. In this paper, video experiments are used to explore the role of cross-domain modelling in building up information about a design. The experiments highlight that cross-domain modelling can be a useful tool to create and structure design information. Findings suggest that consideration of multiple domains encourages discussion during modelling, helps identify design aspects that might otherwise be overlooked, and can help promote consideration of alternative design options. Copyright © 2002-2012 The Design Society. All rights reserved.

The development of a hot rolling process for variable cross-section I-beams

To meet targeted reductions in CO 2 emissions by 2050, demand for metal must be cut, for example through the use of lightweight technologies. However, the efficient production of weight optimized components often requires new, more flexible forming processes. In this paper, a novel hot rolling process is presented for forming I-beams with variable cross-section, which are lighter than prismatic alternatives. First, the new process concept is presented and described. A detailed computational and experimental analysis is then conducted into the capabilities of the process. Results show that the process is capable of producing defect free I-beams with variations in web depth of 30-50%. A full analysis of the process then indicates the likely failure modes, and identifies a safe operating window. Finally, the implications of these results for producing lightweight beams are discussed. © 2012 Elsevier B.V. All rights reserved.

Use of wavefront encoding in optical interconnects and fiber switches for cross talk mitigation

A technique of cross talk mitigation developed for liquid crystal on silicon spatial light modulator based optical interconnects and fiber switches is demonstrated. By purposefully introducing an appropriate aberration into the system, it is possible to reduce the worst-case cross talk by over 10 dB compared to conventional Fourier-transform-based designs. Tests at a wavelength of 674nm validate this approach, and show that there is no noticeable reduction in diffraction efficiency. A 27% spot increase in beam diameter is observed, which is predicted to reduce at longer datacom and telecom wavelengths. © 2012 Optical Society of America.

Improving LVCSR system combination using neural network language model cross adaptation

State-of-the-art large vocabulary continuous speech recognition (LVCSR) systems often combine outputs from multiple subsystems developed at different sites. Cross system adaptation can be used as an alternative to direct hypothesis level combination schemes such as ROVER. The standard approach involves only cross adapting acoustic models. To fully exploit the complimentary features among sub-systems, language model (LM) cross adaptation techniques can be used. Previous research on multi-level n-gram LM cross adaptation is extended to further include the cross adaptation of neural network LMs in this paper. Using this improved LM cross adaptation framework, significant error rate gains of 4.0%-7.1% relative were obtained over acoustic model only cross adaptation when combining a range of Chinese LVCSR sub-systems used in the 2010 and 2011 DARPA GALE evaluations. Copyright © 2011 ISCA.

An analytical model for guided wave inspection optimization for prismatic structures of any cross section.

This paper presents an analytical modeling technique for the simulation of long-range ultrasonic guided waves in structures. The model may be used to predict the displacement field in a prismatic structure arising from any excitation arrangement and may therefore be used as a tool to design new inspection systems. It is computationally efficient and relatively simple to implement, yet gives accuracy similar to finite element analysis and semi-analytical finite element analysis methods. The model has many potential applications; one example is the optimization of part-circumferential arrays where access to the full circumference of the pipe is restricted. The model has been successfully validated by comparison with finite element solutions. Experimental validation has also been carried out using an array of piezoelectric transducer elements to measure the displacement field arising from a single transducer element in an 88.9-mm-diameter pipe. Good agreement has been obtained between the two models and the experimental data.

Language model cross adaptation for LVCSR system combination

State-of-the-art large vocabulary continuous speech recognition (LVCSR) systems often combine outputs from multiple sub-systems that may even be developed at different sites. Cross system adaptation, in which model adaptation is performed using the outputs from another sub-system, can be used as an alternative to hypothesis level combination schemes such as ROVER. Normally cross adaptation is only performed on the acoustic models. However, there are many other levels in LVCSR systems' modelling hierarchy where complimentary features may be exploited, for example, the sub-word and the word level, to further improve cross adaptation based system combination. It is thus interesting to also cross adapt language models (LMs) to capture these additional useful features. In this paper cross adaptation is applied to three forms of language models, a multi-level LM that models both syllable and word sequences, a word level neural network LM, and the linear combination of the two. Significant error rate reductions of 4.0-7.1% relative were obtained over ROVER and acoustic model only cross adaptation when combining a range of Chinese LVCSR sub-systems used in the 2010 and 2011 DARPA GALE evaluations. © 2012 Elsevier Ltd. All rights reserved.