The modified Kanerva model for automatic speech recognition

A parallel processing network derived from Kanerva's associative memory theory Kanerva 1984 is shown to be able to train rapidly on connected speech data and recognize further speech data with a label error rate of 0·68%. This modified Kanerva model can be trained substantially faster than other networks with comparable pattern discrimination properties. Kanerva presented his theory of a self-propagating search in 1984, and showed theoretically that large-scale versions of his model would have powerful pattern matching properties. This paper describes how the design for the modified Kanerva model is derived from Kanerva's original theory. Several designs are tested to discover which form may be implemented fastest while still maintaining versatile recognition performance. A method is developed to deal with the time varying nature of the speech signal by recognizing static patterns together with a fixed quantity of contextual information. In order to recognize speech features in different contexts it is necessary for a network to be able to model disjoint pattern classes. This type of modelling cannot be performed by a single layer of links. Network research was once held back by the inability of single-layer networks to solve this sort of problem, and the lack of a training algorithm for multi-layer networks. Rumelhart, Hinton & Williams 1985 provided one solution by demonstrating the "back propagation" training algorithm for multi-layer networks. A second alternative is used in the modified Kanerva model. A non-linear fixed transformation maps the pattern space into a space of higher dimensionality in which the speech features are linearly separable. A single-layer network may then be used to perform the recognition. The advantage of this solution over the other using multi-layer networks lies in the greater power and speed of the single-layer network training algorithm. © 1989.

Statistical parametric speech synthesis based on speaker and language factorization

An increasingly common scenario in building speech synthesis and recognition systems is training on inhomogeneous data. This paper proposes a new framework for estimating hidden Markov models on data containing both multiple speakers and multiple languages. The proposed framework, speaker and language factorization, attempts to factorize speaker-/language-specific characteristics in the data and then model them using separate transforms. Language-specific factors in the data are represented by transforms based on cluster mean interpolation with cluster-dependent decision trees. Acoustic variations caused by speaker characteristics are handled by transforms based on constrained maximum-likelihood linear regression. Experimental results on statistical parametric speech synthesis show that the proposed framework enables data from multiple speakers in different languages to be used to: train a synthesis system; synthesize speech in a language using speaker characteristics estimated in a different language; and adapt to a new language. © 2012 IEEE.

Phrase-based statistical language generation using graphical models and active learning

Most previous work on trainable language generation has focused on two paradigms: (a) using a statistical model to rank a set of generated utterances, or (b) using statistics to inform the generation decision process. Both approaches rely on the existence of a handcrafted generator, which limits their scalability to new domains. This paper presents BAGEL, a statistical language generator which uses dynamic Bayesian networks to learn from semantically-aligned data produced by 42 untrained annotators. A human evaluation shows that BAGEL can generate natural and informative utterances from unseen inputs in the information presentation domain. Additionally, generation performance on sparse datasets is improved significantly by using certainty-based active learning, yielding ratings close to the human gold standard with a fraction of the data. © 2010 Association for Computational Linguistics.

Graphone model interpolation and Arabic pronunciation generation

This paper extends n-gram graphone model pronunciation generation to use a mixture of such models. This technique is useful when pronunciation data is for a specific variant (or set of variants) of a language, such as for a dialect, and only a small amount of pronunciation dictionary training data for that specific variant is available. The performance of the interpolated n-gram graphone model is evaluated on Arabic phonetic pronunciation generation for words that can't be handled by the Buckwalter Morphological Analyser. The pronunciations produced are also used to train an Arabic broadcast audio speech recognition system. In both cases the interpolated graphone model leads to improved performance. Copyright © 2011 ISCA.

Syllable language models for Mandarin speech recognition: exploiting character language models.

Mandarin Chinese is based on characters which are syllabic in nature and morphological in meaning. All spoken languages have syllabiotactic rules which govern the construction of syllables and their allowed sequences. These constraints are not as restrictive as those learned from word sequences, but they can provide additional useful linguistic information. Hence, it is possible to improve speech recognition performance by appropriately combining these two types of constraints. For the Chinese language considered in this paper, character level language models (LMs) can be used as a first level approximation to allowed syllable sequences. To test this idea, word and character level n-gram LMs were trained on 2.8 billion words (equivalent to 4.3 billion characters) of texts from a wide collection of text sources. Both hypothesis and model based combination techniques were investigated to combine word and character level LMs. Significant character error rate reductions up to 7.3% relative were obtained on a state-of-the-art Mandarin Chinese broadcast audio recognition task using an adapted history dependent multi-level LM that performs a log-linearly combination of character and word level LMs. This supports the hypothesis that character or syllable sequence models are useful for improving Mandarin speech recognition performance.

Application of scalar dissipation rate model to siemens DLE combustors

The standard design process for the Siemens Industrial Turbomachinery, Lincoln, Dry Low Emissions combustion systems has adopted the Eddy Dissipation Model with Finite Rate Chemistry for reacting computational fluid dynamics simulations. The major drawbacks of this model have been the over-prediction of temperature and lack of species data limiting the applicability of the model. A novel combustion model referred to as the Scalar Dissipation Rate Model has been developed recently based on a flamelet type assumption. Previous attempts to adopt the flamelet philosophy with alternative closure models have failed, with the prediction of unphysical phenomenon. The Scalar Dissipation Rate Model (SDRM) was developed from a physical understanding of scalar dissipation rate, signifying the rate of mixing of hot and cold fluids at scales relevant to sustain combustion, in flames and was validated using direct numerical simulations data and experimental measurements. This paper reports on the first industrial application of the SDRM to SITL DLE combustion system. Previous applications have considered ideally premixed laboratory scale flames. The industrial application differs significantly in the complexity of the geometry, unmixedness and operating pressures. The model was implemented into ANSYS-CFX using their inbuilt command language. Simulations were run transiently using Scale Adaptive Simulation turbulence model, which switches between Large Eddy Simulation and Unsteady Reynolds Averaged Navier Stokes using a blending function. The model was validated in a research SITL DLE combustion system prior to being applied to the actual industrial geometry at real operating conditions. This system consists of the SGT-100 burner with a glass square-sectioned combustor allowing for detailed diagnostics. This paper shows the successful validation of the SDRM against time averaged temperature and velocity within measurement errors. The successful validation allowed application of the SDRM to the SGT-100 twin shaft at the relevant full load conditions. Limited validation data was available due to the complexity of measurement in the real geometry. Comparison of surface temperatures and combustor exit temperature profiles showed an improvement compared to EDM/FRC model. Furthermore, no unphysical phenomena were predicted. This paper presents the successful application of the SDRM to the industrial combustion system. The model shows a marked improvement in the prediction of temperature over the EDM/FRC model previously used. This is of significant importance in the future applications of combustion CFD for understanding of hardware mechanical integrity, combustion emissions and dynamics of the flame. Copyright © 2012 by ASME.

Automatic construction and natural-language description of nonparametric regression models

Copyright © 2014, Association for the Advancement of Artificial Intelligence (www.aaai.org). All rights reserved. This paper presents the beginnings of an automatic statistician, focusing on regression problems. Our system explores an open-ended space of statistical models to discover a good explanation of a data set, and then produces a detailed report with figures and natural- language text. Our approach treats unknown regression functions non- parametrically using Gaussian processes, which has two important consequences. First, Gaussian processes can model functions in terms of high-level properties (e.g. smoothness, trends, periodicity, changepoints). Taken together with the compositional structure of our language of models this allows us to automatically describe functions in simple terms. Second, the use of flexible nonparametric models and a rich language for composing them in an open-ended manner also results in state- of-the-art extrapolation performance evaluated over 13 real time series data sets from various domains.

A literature and practice review to develop sustainable business model archetypes

Eco-innovations, eco-efficiency and corporate social responsibility practices define much of the current industrial sustainability agenda. While important, they are insufficient in themselves to deliver the holistic changes necessary to achieve long-term social and environmental sustainability. How can we encourage corporate innovation that significantly changes the way companies operate to ensure greater sustainability? Sustainable business models (SBM) incorporate a triple bottom line approach and consider a wide range of stakeholder interests, including environment and society. They are important in driving and implementing corporate innovation for sustainability, can help embed sustainability into business purpose and processes, and serve as a key driver of competitive advantage. Many innovative approaches may contribute to delivering sustainability through business models, but have not been collated under a unifying theme of business model innovation. The literature and business practice review has identified a wide range of examples of mechanisms and solutions that can contribute to business model innovation for sustainability. The examples were collated and analysed to identify defining patterns and attributes that might facilitate categorisation. Sustainable business model archetypes are introduced to describe groupings of mechanisms and solutions that may contribute to building up the business model for sustainability. The aim of these archetypes is to develop a common language that can be used to accelerate the development of sustainable business models in research and practice. The archetypes are: Maximise material and energy efficiency; Create value from 'waste'; Substitute with renewables and natural processes; Deliver functionality rather than ownership; Adopt a stewardship role; Encourage sufficiency; Re-purpose the business for society/environment; and Develop scale-up solutions. © 2014 The Authors. Published by Elsevier Ltd. All rights reserved.

Computational Consequences of Agreement and Ambiguity in Natural Language

The computer science technique of computational complexity analysis can provide powerful insights into the algorithm-neutral analysis of information processing tasks. Here we show that a simple, theory-neutral linguistic model of syntactic agreement and ambiguity demonstrates that natural language parsing may be computationally intractable. Significantly, we show that it may be syntactic features rather than rules that can cause this difficulty. Informally, human languages and the computationally intractable Satisfiability (SAT) problem share two costly computional mechanisms: both enforce agreement among symbols across unbounded distances (Subject-Verb agreement) and both allow ambiguity (is a word a Noun or a Verb?).

Formalizing Triggers: A Learning Model for Finite Spaces

In a recent seminal paper, Gibson and Wexler (1993) take important steps to formalizing the notion of language learning in a (finite) space whose grammars are characterized by a finite number of parameters. They introduce the Triggering Learning Algorithm (TLA) and show that even in finite space convergence may be a problem due to local maxima. In this paper we explicitly formalize learning in finite parameter space as a Markov structure whose states are parameter settings. We show that this captures the dynamics of TLA completely and allows us to explicitly compute the rates of convergence for TLA and other variants of TLA e.g. random walk. Also included in the paper are a corrected version of GW's central convergence proof, a list of "problem states" in addition to local maxima, and batch and PAC-style learning bounds for the model.

A Computational Model for the Acquisition and Use of Phonological Knowledge

Does knowledge of language consist of symbolic rules? How do children learn and use their linguistic knowledge? To elucidate these questions, we present a computational model that acquires phonological knowledge from a corpus of common English nouns and verbs. In our model the phonological knowledge is encapsulated as boolean constraints operating on classical linguistic representations of speech sounds in term of distinctive features. The learning algorithm compiles a corpus of words into increasingly sophisticated constraints. The algorithm is incremental, greedy, and fast. It yields one-shot learning of phonological constraints from a few examples. Our system exhibits behavior similar to that of young children learning phonological knowledge. As a bonus the constraints can be interpreted as classical linguistic rules. The computational model can be implemented by a surprisingly simple hardware mechanism. Our mechanism also sheds light on a fundamental AI question: How are signals related to symbols?

Naive Physics, Event Perception, Lexical Semantics, and Language Acquisition

This thesis proposes a computational model of how children may come to learn the meanings of words in their native language. The proposed model is divided into two separate components. One component produces semantic descriptions of visually observed events while the other correlates those descriptions with co-occurring descriptions of those events in natural language. The first part of this thesis describes three implementations of the correlation process whereby representations of the meanings of whole utterances can be decomposed into fragments assigned as representations of the meanings of individual words. The second part of this thesis describes an implemented computer program that recognizes the occurrence of simple spatial motion events in simulated video input.

AMAR: A Computational Model of Autosegmental Phonology

This report describes a computational system with which phonologists may describe a natural language in terms of autosegmental phonology, currently the most advanced theory pertaining to the sound systems of human languages. This system allows linguists to easily test autosegmental hypotheses against a large corpus of data. The system was designed primarily with tonal systems in mind, but also provides support for tree or feature matrix representation of phonemes (as in The Sound Pattern of English), as well as syllable structures and other aspects of phonological theory. Underspecification is allowed, and trees may be specified before, during, and after rule application. The association convention is automatically applied, and other principles such as the conjunctivity condition are supported. The method of representation was designed such that rules are designated in as close a fashion as possible to the existing conventions of autosegmental theory while adhering to a textual constraint for maximum portability.