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.

Transposon mutagenesis in a hyper-invasive clinical isolate of Campylobacter jejuni reveals a number of genes with potential roles in invasion.

Transposon mutagenesis has been applied to a hyper-invasive clinical isolate of Campylobacter jejuni, 01/51. A random transposon mutant library was screened in an in vitro assay of invasion and 26 mutants with a significant reduction in invasion were identified. Given that the invasion potential of C. jejuni is relatively poor compared to other enteric pathogens, the use of a hyper-invasive strain was advantageous as it greatly facilitated the identification of mutants with reduced invasion. The location of the transposon insertion in 23 of these mutants has been determined; all but three of the insertions are in genes also present in the genome-sequenced strain NCTC 11168. Eight of the mutants contain transposon insertions in one region of the genome (approximately 14 kb), which when compared with the genome of NCTC 11168 overlaps with one of the previously reported plasticity regions and is likely to be involved in genomic variation between strains. Further characterization of one of the mutants within this region has identified a gene that might be involved in adhesion to host cells.