Development of the 2003 CU-HTK conversational telephone speech transcription system

This paper describes the development of the 2003 CU-HTK large vocabulary speech recognition system for Conversational Telephone Speech (CTS). The system was designed based on a multi-pass, multi-branch structure where the output of all branches is combined using system combination. A number of advanced modelling techniques such as Speaker Adaptive Training, Heteroscedastic Linear Discriminant Analysis, Minimum Phone Error estimation and specially constructed Single Pronunciation dictionaries were employed. The effectiveness of each of these techniques and their potential contribution to the result of system combination was evaluated in the framework of a state-of-the-art LVCSR system with sophisticated adaptation. The final 2003 CU-HTK CTS system constructed from some of these models is described and its performance on the DARPA/NIST 2003 Rich Transcription (RT-03) evaluation test set is discussed.

Unsupervised training for mandarin broadcast news and conversation transcription

A significant cost in obtaining acoustic training data is the generation of accurate transcriptions. For some sources close-caption data is available. This allows the use of lightly-supervised training techniques. However, for some sources and languages close-caption is not available. In these cases unsupervised training techniques must be used. This paper examines the use of unsupervised techniques for discriminative training. In unsupervised training automatic transcriptions from a recognition system are used for training. As these transcriptions may be errorful data selection may be useful. Two forms of selection are described, one to remove non-target language shows, the other to remove segments with low confidence. Experiments were carried out on a Mandarin transcriptions task. Two types of test data were considered, Broadcast News (BN) and Broadcast Conversations (BC). Results show that the gains from unsupervised discriminative training are highly dependent on the accuracy of the automatic transcriptions. © 2007 IEEE.

The role of androgens in fetal growth: observational study in two genetic models of disordered androgen signalling.

OBJECTIVE: To examine the role of androgens on birth weight in genetic models of altered androgen signalling. SETTING: Cambridge Disorders of Sex Development (DSD) database and the Swedish national screening programme for congenital adrenal hyperplasia (CAH). PATIENTS: (1) 29 girls with XY karyotype and mutation positive complete androgen insensitivity syndrome (CAIS); (2) 43 girls and 30 boys with genotype confirmed CAH. MAIN OUTCOME MEASURES: Birth weight, birth weight-for-gestational-age (birth weight standard deviation score (SDS)) calculated by comparison with national references. RESULTS: Mean birth weight SDS in CAIS XY infants was higher than the reference for girls (mean, 95% CI: 0.4, 0.1 to 0.7; p=0.02) and was similar to the national reference for boys (0.1, -0.2 to 0.4). Birth weight SDS in CAH girls was similar to the national reference for girls (0.0, -0.2 to 0.2) and did not vary by severity of gene mutation. Birth weight SDS in CAH boys was also similar to the national reference for boys (0.2, -0.2 to 0.6). CONCLUSION: CAIS XY infants have a birth weight distribution similar to normal male infants and birth weight is not increased in infants with CAH. Alterations in androgen signalling have little impact on birth weight. Sex dimorphism in birth size is unrelated to prenatal androgen exposure.

A species conserving genetic algorithm for multimodal function optimization

This paper introduces a new technique called species conservation for evolving parallel subpopulations. The technique is based on the concept of dividing the population into several species according to their similarity. Each of these species is built around a dominating individual called the species seed. Species seeds found in the current generation are saved (conserved) by moving them into the next generation. Our technique has proved to be very effective in finding multiple solutions of multimodal optimization problems. We demonstrate this by applying it to a set of test problems, including some problems known to be deceptive to genetic algorithms.