A speech-music discriminator using HILN model based features

We propose a simple speech music discriminator that uses features based on HILN(Harmonics, Individual Lines and Noise) model. We have been able to test the strength of the feature set on a standard database of 66 files and get an accuracy of around 97%. We also have tested on sung queries and polyphonic music and have got very good results. The current algorithm is being used to discriminate between sung queries and played (using an instrument like flute) queries for a Query by Humming(QBH) system currently under development in the lab.

Improving outcome of childhood bacterial meningitis by simplified treatment : Experience from Angola

Background Acute bacterial meningitis (BM) continues to be an important cause of childhood mortality and morbidity, especially in developing countries. Prognostic scales and the identification of risk factors for adverse outcome both aid in assessing disease severity. New antimicrobial agents or adjunctive treatments - except for oral glycerol - have essentially failed to improve BM prognosis. A retrospective observational analysis found paracetamol beneficial in adult bacteraemic patients, and some experts recommend slow β-lactam infusion. We examined these treatments in a prospective, double-blind, placebo-controlled clinical trial. Patients and methods A retrospective analysis included 555 children treated for BM in 2004 in the infectious disease ward of the Paediatric Hospital of Luanda, Angola. Our prospective study randomised 723 children into four groups, to receive a combination of cefotaxime infusion or boluses every 6 hours for the first 24 hours and oral paracetamol or placebo for 48 hours. The primary endpoints were 1) death or severe neurological sequelae (SeNeSe), and 2) deafness. Results In the retrospective study, the mortality of children with blood transfusion was 23% (30 of 128) vs. without blood transfusion 39% (109 of 282; p=0.004). In the prospective study, 272 (38%) of the children died. Of those 451 surviving, 68 (15%) showed SeNeSe, and 12% (45 of 374) were deaf. Whereas no difference between treatment groups was observable in primary endpoints, the early mortality in the infusion-paracetamol group was lower, with the difference (Fisher s exact test) from the other groups at 24, 48, and 72 hours being significant (p=0.041, 0.0005, and 0.005, respectively). Prognostic factors for adverse outcomes were impaired consciousness, dyspnoea, seizures, delayed presentation, and absence of electricity at home (Simple Luanda Scale, SLS); the Bayesian Luanda Scale (BLS) also included abnormally low or high blood glucose. Conclusions New studies concerning the possible beneficial effect of blood transfusion, and concerning longer treatment with cefotaxime infusion and oral paracetamol, and a study to validate our simple prognostic scales are warranted.