Growth setbacks in new firms

New firms in emerging industries are subject to complex dynamic processes which defy the attempts at prediction embodied in business conjectures. Discontinuous growth is common, but the issue of interruptions in the early growth of new firms has not been adequately addressed in the mainstream literature. We examine the prevalence of interruptions to growth in a cohort study of the growth trajectories of firms founded in 1990, then look to cases studies of individual firms to investigate underlying causes. We find that substantial growth is rare and continuous growth unusual, and that growth interruptions are the result of both internal and external dynamics. The managers of growing firms face shortages of vital resources and significant problems of resource synchronisation and coordination, many of which can lead to what are, in effect, changes of phase state. Meanwhile, the volatile environment of an emerging industry presents particular problems to young firms which have not yet built up reserves to sustain them through short-term crises. However, problem solving by those that survive provides an important source of learning which can underpin their future development. © 2004 Elsevier Ltd. All rights reserved.

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.