Introduction: Background: Optimum glycemic control is pivotal to prevent complications and premature mortality, and improve quality of life in patients with type 2 diabetes mellitus (T2DM) Patients uncontrolled on mono- or dual- antidiabetic therapy require triple drug therapy (TDT) to achieve optimal glycemic control Objectives: Objective: This survey was carried out to understand the clinical utilization of TDT in attaining durable glycemic control in patients with T2DM Methods: Methods: A structured objective questionnaire was used for the survey and responses were sought from 270 Healthcare practitioners (HCPs) across India during May 15 to May 30, 2020 via Google forms The questionnaire consisted of questions related to proportion of patients with uncontrolled glycemia, role and choice of TDT in achieving durable glycemic control, importance of cost of therapy, etc The responses were analyzed descriptively Results: Results: The respondent HCP pool comprised consulting physicians (47%), diabetologists (30%), endocrinologists (16%), and family physicians (7%) More than 45% of HCPs agreed that 30-50% of their diabetes patients have HbA1c >8 5%, and uncontrolled fasting and postprandial blood glucose levels while 21% mentioned such uncontrolled glycemia in more than 50% of their patients Approx 52% of the HCPs think that 50-70% of their patients can be brought under control with TDT, while a quarter of HCPs believe that more than 70% of their patients can be effectively managed with TDT 72% of respondents aligned in their views that the cost of anti-diabetic therapy is an important parameter during Covid-19 pandemic In uncontrolled T2DM patients taking a combination of modern sulphonylurea and metformin, HCPs rated DPP-4 inhibitor, pioglitazone and voglibose as third line options for achieving durable glycemic control 60% of the HCPs opined that triple drug therapy can delay the initiation of insulin therapy in patients with T2DM Almost two-thirds of the HCPs agreed that double and triple fixed dose OAD combinations have better pharmacy availability than insulin, enabling better patient accessibility Conclusion: Conclusion: A high proportion of patients have a suboptimal glycemic control during covid-19 pandemic, and TDT can control a majority of these patients The survey also highlights the HCPs' opinion that the triple drug therapy has multiple patient-centric benefits like cost effectiveness, accessibility, delaying insulin usage

In the IARPA MATERIAL program, information retrieval (IR) is treated as a hard detection problem; the system has to output a single global ranking over all queries, and apply a hard threshold on this global list to come up with all the hypothesized relevant documents. This means that how queries are ranked relative to each other can have a dramatic impact on performance. In this paper, we study such a performance measure, the Average Query Weighted Value (AQWV), which is a combination of miss and false alarm rates. AQWV requires that the same detection threshold is applied to all queries. Hence, detection scores of different queries should be comparable, and, to do that, a score normalization technique (commonly used in keyword spotting from speech) should be used. We describe unsupervised methods for score normalization, which are borrowed from the speech field and adapted accordingly for IR, and demonstrate that they greatly improve AQWV on the task of cross-language information retrieval (CLIR), on three low-resource languages used in MATERIAL. We also present a novel supervised score normalization approach which gives additional gains.

We describe the human triage scenario envisioned in the Cross-Lingual Information Retrieval (CLIR) problem of the [REDUCT] Program. The overall goal is to maximize the quality of the set of documents that is given to a bilingual analyst, as measured by the AQWV score. The initial set of source documents that are retrieved by the CLIR system is summarized in English and presented to human judges who attempt to remove the irrelevant documents (false alarms); the resulting documents are then presented to the analyst. First, we describe the AQWV performance measure and show that, in our experience, if the acceptance threshold of the CLIR component has been optimized to maximize AQWV, the loss in AQWV due to false alarms is relatively constant across many conditions, which also limits the possible gain that can be achieved by any post filter (such as human judgments) that removes false alarms. Second, we analyze the likely benefits for the triage operation as a function of the initial CLIR AQWV score and the ability of the human judges to remove false alarms without removing relevant documents. Third, we demonstrate that we can increase the benefit for human judgments by combining the human judgment scores with the original document scores returned by the automatic CLIR system.

In this paper, we describe a cross-lingual information retrieval (CLIR) system that, given a query in English, and a set of audio and text documents in a foreign language, can return a scored list of relevant documents, and present findings in a summary form in English. Foreign audio documents are first transcribed by a state-of-the-art pretrained multilingual speech recognition model that is finetuned to the target language. For text documents, we use multiple multilingual neural machine translation (MT) models to achieve good translation results, especially for low/medium resource languages. The processed documents and queries are then scored using a probabilistic CLIR model that makes use of the probability of translation from GIZA translation tables and scores from a Neural Network Lexical Translation Model (NNLTM). Additionally, advanced score normalization, combination, and thresholding schemes are employed to maximize the Average Query Weighted Value (AQWV) scores. The CLIR output, together with multiple translation renderings, are selected and translated into English snippets via a summarization model. Our turnkey system is language agnostic and can be quickly trained for a new low-resource language in few days.

We propose a neural network model to estimate word translation probabilities for Cross-Lingual Information Retrieval (CLIR). The model estimates better probabilities for word translations than automatic word alignments alone, and generalizes to unseen source-target word pairs. We further improve the lexical neural translation model (and subsequently CLIR), by incorporating source word context, and by encoding the character sequences of input source words to generate translations of out-of-vocabulary words. To be effective, neural network models typically need training on large amounts of data labeled directly on the final task, in this case relevance to queries. In contrast, our approach only requires parallel data to train the translation model, and uses an unsupervised model to compute CLIR relevance scores. We report results on the retrieval of text and speech documents from three morphologically complex languages with limited training data resources (Swahili, Tagalog, and Somali) and short English queries. Despite training on only about 2M words of parallel training data for each language, we obtain neural network translation models that are very effective for this task. We also obtain further improvements using (i) a modified relevance model, which uses the probability of occurrence of a translation of each query term in the source document, and (ii) confusion networks (instead of 1-best output) that encode multiple transcription alternatives in the output of an Automatic Speech Recognition (ASR) system. We achieve overall MAP relative improvements of up to 24% on Swahili, 50% on Tagalog, and 39% on Somali over the baseline probabilistic model, and larger improvements over monolingual retrieval from machine translation output.

Ship classification is the task of automatically classifying a ship into one of several predefined categories using features that are derived from a collected underwater sonar signal. In this paper, we follow a cross-disciplinary approach, where tools and methodologies from speech processing are used to tackle this problem. The corpus that we use was collected by the Scripps Institution of Oceanography, UCSD, with a single-hydrophone setup, over the course of nine years. Of that corpus, we use data from 1861 ships for our research. We present our methodology for feature extraction, ship detection, and the classification of ships into a number of ship types. To demonstrate the effectiveness of our approach, we compare our results with an implementation based on the method in [1]. We also report on a series of controlled experiments that measure the effect of different parameter settings in the feature generation and classification stages.

Recent work has shown success in using neural network language models (NNLMs) as features in MT systems. Here, we present a novel formulation for a neural network joint model (NNJM), which augments the NNLM with a source context window. Our model is purely lexicalized and can be integrated into any MT decoder. We also present several variations of the NNJM which provide significant additive improvements.Although the model is quite simple, it yields strong empirical results. On the NIST OpenMT12 Arabic-English condition, the NNJM features produce a gain of +3.0 BLEU on top of a powerful, featurerich baseline which already includes a target-only NNLM. The NNJM features also produce a gain of +6.3 BLEU on top of a simpler baseline equivalent to Chiang's (2007) original Hiero implementation.Additionally, we describe two novel techniques for overcoming the historically high cost of using NNLM-style models in MT decoding. These techniques speed up NNJM computation by a factor of 10,000x, making the model as fast as a standard back-off LM.

In our recent work, we have proposed that multiple behavior demonstrations can be automatically combined to generate an Example-Tracing Tutor model. In this paper, we compare four algorithms for this problem using a number of different metrics for two different datasets, one of which is publicly available. Our experiments show that these four algorithms are complementary to each other in terms of their performance along the different metrics. These findings make a case for incorporating multiple algorithms for building behavior graphs into authoring tools for Intelligent Tutoring Systems (ITS) that use behavior graphs.

We present a systematic study of the effect of crowdsourced translations on Machine Translation performance. We compare Machine Translation systems trained on the same data but with translations obtained using Amazon’s Mechanical Turk vs. professional translations, and show that the same performance is obtained from Mechanical Turk translations at 1/5th the cost. We also show that adding a Mechanical Turk reference translation of the development set improves parameter tuning and output evaluation.

This article presents several techniques for integrating information from a rule-based machine translation (RBMT) system into a statistical machine translation (SMT) framework. These techniques are grouped into three parts that correspond to the type of information integrated: the morphological, lexical, and system levels. The first part presents techniques that use information from a rule-based morphological tagger to do morpheme splitting of the Arabic source text. We also compare with the results of using a statistical morphological tagger. In the second part, we present two ways of using Arabic diacritics to improve SMT results, both based on binary decision trees. The third part presents a system combination method that combines the outputs of the RBMT and the SMT systems, leveraging the strength of each. This article shows how language specific information obtained through a deterministic rule-based process can be used to improve SMT, which is mostly language-independent.

We describe several improvements that we have made in the BBN BYBLOS OCR System. First, we adopted continuous density hidden Markov models (HMMs) rather than discrete density HMMs. This resulted in improved accuracy when more training data is available. It also allowed us to use unsupervised speaker adaptation algorithms (borrowed from speech recognition) for adaptation to font, style, and quality. Second. we sped up the character recognition by a factor of about 50 so that a full page of 2,000 characters requires about 30 to 40 seconds for processing. Third, we tested the system on Chinese characters. This required development of tools to create a training corpus from available sources. It also required techniques for dealing with an open set of characters, where some of the characters may have no real training data. The end result was 1.2% character error on newspaper data.

: This is the final report for Year 3 of the GALE project, whose objective is to transcribe and translate foreign spoken and written languages into English and to distill the transcription into accurate information for use by our military. Below, we summarize the work performed by the BBN-led AGILE Team in Year 3. A more detailed description of the work performed can be found in the DARPA/IPTO Quarterly Status Reports for this project. The Appendix contains the accomplishments of three additional efforts: Serif Maturation, Broadcast Monitoring System One-Year Archive, Robust Automatic Transcription of Speech (RATS) and Serif Research.

This paper presents a set of experiments that we conducted in order to optimize the performance of an Arabic/English machine translation system on broadcast news and conversational speech data. Proper integration of speech-to-text (STT) and machine translation (MT) requires special attention to issues such as sentence boundary detection, punctuation, STT accuracy, tokenization, conversion of spoken numbers and dates to written form, optimization of MT decoding weights, and scoring. We discuss these issues, and show that a carefully tuned STT/MT integration can lead to significant translation accuracy improvements compared to simply feeding the regular STT output to a text MT system.

This paper investigates the use of several language model adaptation techniques applied to the task of machine translation from Arabic broadcast speech. Unsupervised and discriminative approaches slightly outperform the traditional perplexity-based optimization technique. Language model adaptation, when used for n-best rescoring, improves machine translation performance by 0.3-0.4 BLEU and reduces translation edit rate (TER) by 0.2-0.5% compared to an unadapted LM.

This paper describes the progress made in the transcription of broadcast news (BN) and conversational telephone speech (CTS) within the combined BBN/LIMSI system from May 2002 to September 2004. During that period, BBN and LIMSI collaborated in an effort to produce significant reductions in the word error rate (WER), as directed by the aggressive goals of the Effective, Affordable, Reusable, Speech-to-text [Defense Advanced Research Projects Agency (DARPA) EARS] program. The paper focuses on general modeling techniques that led to recognition accuracy improvements, as well as engineering approaches that enabled efficient use of large amounts of training data and fast decoding architectures. Special attention is given on efforts to integrate components of the BBN and LIMSI systems, discussing the tradeoff between speed and accuracy for various system combination strategies. Results on the EARS progress test sets show that the combined BBN/LIMSI system achieved relative reductions of 47% and 51% on the BN and CTS domains, respectively.

In the official scoring, conducted by NIST, BBN's HARC system 3. produced a weighted SLS score of 43.7 on all 687 evaluable utterances in the test set. This was the lowest error achieved by any of the 7 systems evaluated. 4. For the SPREC evaluation BBN's BYBLOS system achieved a word error rate of 6.2% on the same 687 utterances and 9.4% on the entire test set of 971 utterances. These results were significantly better than any other speech system evaluated.

We examine a new, intuitive measure for evaluating machine-translation output that avoids the knowledge intensiveness of more meaning-based approaches, and the labor-intensiveness of human judgments. Translation Edit Rate (TER) measures the amount of editing that a human would have to perform to change a system output so it exactly matches a reference translation. We show that the single-reference variant of TER correlates as well with human judgments of MT quality as the four-reference variant of BLEU. We also define a human-targeted TER (or HTER) and show that it yields higher correlations with human judgments than BLEU—even when BLEU is given human-targeted references. Our results indicate that HTER correlates with human judgments better than HMETEOR and that the four-reference variants of TER and HTER correlate with human judgments as well as—or better than—a second human judgment does.

Summary form only given. The state of the art in automatic speech recognition has reached the point that searching for and extracting information from large speech repositories or streaming audio has become a growing reality. This paper summarizes the technologies that have been instrumental in making audio as searchable as text, including speech recognition, speaker clustering, segmentation, and identification; topic classification; and story segmentation. Once speech is turned into text, information extraction methods can then be applied, such as named entity extraction, finding relationships between named entities, and resolution of anaphoric references. Examples of deployed systems for information extraction from speech, which incorporate some of the aforementioned technologies, will be given.