Classifier performance is often enhanced through combining multiple streams of information. In the context of multistream HMM/ANN systems in ASR, a confidence measure widely used in classifier combination is the entropy of the posteriors distribution output from each ANN, which generally increases as classification becomes less reliable. The rule most commonly used is to select the ANN with the minimum entropy. However, this is not necessarily the best way to use entropy in classifier combination. In this article, we test three new entropy based combination rules in a fullcombination multi-stream HMM/ANN system for noise robust speech recognition. Best results were obtained by combining all the classifiers having entropy below average using a weighting proportional to their inverse entropy.
Abstract. In general, entropy gives us a measure of the number of bits required to represent some information. When applied to probability mass function (PMF), entropy can also be used to measure the "peakiness" of a distribution. In this paper, we propose using the entropy of short time Fourier transform spectrum, normalised as PMF, as an additional feature for automatic speech recognition (ASR). It is indeed expected that a peaky spectrum, representation of clear formant structure in the case of voiced sounds, will have low entropy, while a flatter spectrum corresponding to non-speech or noisy regions will have higher entropy. Extending this reasoning further, we introduce the idea of multi-band/multi-resolution entropy feature where we divide the spectrum into equal size sub-bands and compute entropy in each sub-band. The results presented in this paper show that multi-band entropy features used in conjunction with normal cepstral features improve the performance of ASR system. 2 IDIAP-RR 03-56
In this paper, the task of text segmentation is approached from a topic modeling perspective. We investigate the use of latent Dirichlet allocation (LDA) topic model to segment a text into semantically coherent segments. A major benefit of the proposed approach is that along with the segment boundaries, it outputs the topic distribution associated with each segment. This information is of potential use in applications like segment retrieval and discourse analysis. The new approach outperforms a standard baseline method and yields significantly better performance than most of the available unsupervised methods on a benchmark dataset.
PEGylated carboxyhemoglobin bovine (SANGUINATE) is a dual action carbon monoxide releasing (CO)/oxygen (O2 ) transfer agent for the treatment of hypoxia. Its components inhibit vasoconstriction, decrease extravasation, limit reactive oxygen species production, enhance blood rheology, and deliver oxygen to the tissues. Animal models of cerebral ischemia, peripheral ischemia, and myocardial ischemia demonstrated SANGUINATE's efficacy in reducing myocardial infarct size, limiting necrosis from cerebral ischemia, and promoting more rapid recovery from hind limb ischemia. In a Phase I trial, three cohorts of eight healthy volunteers received single ascending doses of 80, 120, or 160 mg/kg of SANGUINATE. Two volunteers within each cohort served as a saline control. There were no serious adverse events. Serum haptoglobin decreased, but did not appear to be dose related. The T1/2 was dose dependent and ranged from 7.9 to 13.8 h. In addition to the Phase I trial, SANGUINATE was used under an expanded access emergency Investigational New Drug. SANGUINATE was found to be safe and well tolerated in a Phase I clinical trial, and therefore it will advance into further clinical trials in patients.
In this paper, the task of text segmentation is approached from a topic modeling perspective. We investigate the use of two unsupervised topic models, latent Dirichlet allocation (LDA) and multinomial mixture (MM), to segment a text into semantically coherent parts. The proposed topic model based approaches consistently outperform a standard baseline method on several datasets. A major benefit of the proposed LDA based approach is that along with the segment boundaries, it outputs the topic distribution associated with each segment. This information is of potential use in applications such as segment retrieval and discourse analysis. However, the proposed approaches, especially the LDA based method, have high computational requirements. Based on an analysis of the dynamic programming (DP) algorithm typically used for segmentation, we suggest a modification to DP that dramatically speeds up the process with no loss in performance. The proposed modification to the DP algorithm is not specific to the topic models only; it is applicable to all the algorithms that use DP for the task of text segmentation.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
hi@scite.ai
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.