Search

Advanced heart failure, an end-stage disease characterized by high mortality and morbidity despite standard medical therapy, requires various therapeutic strategies like heart transplant and long-term mechanical circulatory support. Echocardiography is the main imaging technique to identify transitions to advanced stages of disease and guide risk stratification and therapeutic decision-making processes. Progressive development of advanced echocardiographic techniques allows more comprehensive assessment of the hemodynamic and structural profiles of patients with advanced heart failure, and its use in clinical practice continues to expand. This article provides an overview of basic and emerging echocardiographic tools to assess patients with advanced heart failure.

Background Hypertrophic cardiomyopathy (HCM) is characterized by extremely varied phenotypic expression ranging from asymptomatic to heart failure (HF) to sudden cardiac death (SCD). Although children with HCM are considered in the highest risk spectrum, the most common recommendations on pharmacological and non-pharmacological treatment (i.e. drugs, ICD, septal reduction procedures, inclusion in cardiac transplantation list) are often disregarded or too much postponed in this setting and strong evidence-based risk prediction models are missing. A systematic cardiopulmonary exercise test (CPET) assessment might be helpful to disclose an unsuspected functional limitation. Purpose The aim of our multicenter retrospective study was to investigate possible clinical insights, in terms of functional and prognostic assessment, coming from a full CPET assessment in a cohort of pediatric HCM outpatients aged less than 18 years old. Methods Sixty consecutive pediatric HCM outpatients aged Results An impaired exercise capacity, consisting on peak VO2 values Respect to the control Group, the HCM patients showed a significantly poorer functional status in terms of maximum workload achieved, peak VO2 (regardless the adopted correction), circulatory power and VE/VCO2 slope values (Figure 1, panel A). HCM patients who experience adverse events during the follow-up (Event Group) showed the worst CPET profile (Figure 1, panel B). The composite end-point occurred more frequently in patients with the worst CPETs' profiles. At the univariate analysis, peak VO2% was the variable with the strongest association with adverse events at follow-up (C-index=0.72, p=0.025) and a cut-off value equal to 60% was the most accurate in identifying those patients at the highest risk (Figure 2). Conclusions Our findings support the role of CPET analysis as an insightful approach in the young HCM clinical management. In a group of young asymptomatic or slightly symptomatic HCM patients, the CPET allowed us to estimate accurately their functional capacity and to disclose a portion of un-recognized exercise impairment. Our data argue in favor of a possible role of some CPET-derived variables in the early identification of those young HCM patients at highest risk of HCM related events. Funding Acknowledgement Type of funding sources: None. Figure 1Figure 2

The usefulness of cardiopulmonary exercise test (CPET) in adult hypertrophic cardiomyopathy (HCM) patients is well-known, whereas its role in pediatric HCM patients has not yet been explored. The present study investigates possible insights from a CPET assessment in a cohort of pediatric HCM outpatients in terms of functional and prognostic assessment. Sixty consecutive pediatric HCM outpatients aged &lt, 18 years old were enrolled, each of them undergoing a full clinical assessment including a CPET, a group of 60 healthy subjects served as controls. A unique composite end-point of heart failure (HF) related and sudden cardiac death (SCD) or SCD-equivalent events was also explored. During a median follow-up of 53 months (25th–75th: 13–84 months), a total of 13 HF- and 7 SCD-related first events were collected. Compared to controls, HCM patients showed an impaired functional capacity with most of them showing peak oxygen uptake (pVO2) values of &lt, 80% of the predicted, clearly discrepant with functional New York Heart Association class assessment. The composite end-point occurred more frequently in patients with the worst CPETs’ profiles. At the univariate analysis, pVO2% was the variable with the strongest association with adverse events at follow-up (C-index = 0.72, p = 0.025) and a cut-off value equal to 60% was the most accurate in identifying those patients at the highest risk. In a pediatric HCM subset, the CPET assessment allows a true functional capacity estimation and it might be helpful in identifying early those patients at high risk of events.

Funding Acknowledgements Type of funding sources: None. Background Intermittent infusions of levosimendan in an outpatient setting have been associated with improved symptoms and reduced hospitalizations in patients with advanced heart failure (HF). Little is known on the potential effect of intermittent levosimendan infusions on right ventricle (RV)-pulmonary arterial (PA) coupling in ambulatory patients with advanced HF. Purpose Aim of the present study was to explore the effects of intermittent levosimendan infusions on the ratio between tricuspid annular plane systolic excursion (TAPSE) and pulmonary artery systolic pressure (PAPs), an echocardiographic measure of RV/PA coupling, in ambulatory patients with advanced HF, and on left ventricular arterial coupling (VAC), expressed as the relationship between arterial elastance (Ea) and ventricular elastance (Ees). Methods 17 ambulatory patients with advanced HF treated with intermittent levosimendan (6-hour intravenous infusion, 0.2 ug/kg/min without bolus) received baseline clinical, biochemical and echocardiographic evaluation and changes in TAPSE/PAPs ratio were assessed from baseline to 48 hours after the infusion, based on the pharmacokinetic profile of levosimendan. VAC, expressed as Ea/Ees ratio, was obtained by a calculator (iElastance) designed for non-invasive single beat measure of end-systolic Ees and Ea according to Chen"s method. Results After 48 hours from levosimendan infusion, there was a significant improvement of TAPSE/PAPs ratio (p = 0.04), stroke voume (SV) (p = 0.05) and cardiac output (CO)(p = 0.04). We observed a significant reduction of Ea (p = 0.007) and of Ees (p = 0.01) and a non significant improvement of VAC (p = 0.4)(Tab.1). Conclusion. Our results show that an intermittent 6-hour levosimendan infusion at 0.2 ug/kg/min improves after 48 hours RV-PA coupling, SV, CO, Ea and Ees in ambulatory patients with advanced HF. Further studies including more patients are necessary to confirm these preliminary findings. Tab.1 Baseline 48h afterlevosimendan infusion p value NT-proBNP (pg/mL) 5607 ± 4300 3868 ± 3856

Despite progressive improvement in medical therapy and standard care, the exercise capacity of heart transplant recipients is reduced compared with age-matched healthy individuals. Exercise-based rehabilitation programs have been shown to improve the exercise capacity of transplant patients through a multifactorial effect. In this context, high-intensity interval exercise is a growing field of research, with current evidence suggesting a major benefit in heart transplant recipients compared with a conventional training protocol. Therefore, this study aimed to provide an overview of the mechanisms involved in the reduced exercise capacity of heart transplant patients and a review of current rehabilitation strategies with a special focus on the mechanisms and clinical effects of high-intensity interval training exercise.

Hypertriglyceridemia (triglycerides ≥200 mg/dL) is a major cardiovascular risk factor. Despite the high incidence of this condition in the Italian population, epidemiological information remains limited. To analyze a large database of clinical charts from general practitioners in the Rome area of Italy. In this study, the DATAMEG database of patients treated in the Rome area between 2000 and 2015 was analyzed. The database was searched for information on (1) all patients diagnosed with hypertriglyceridemia, (2) all prescriptions for the treatment of hypertriglyceridemia, and (3) all patients who had an acute myocardial infarction. The overall prevalence of hypertriglyceridemia was 4.4% (3647/82,595). Among patients followed from January 1, 2015 onwards, 2786/55,345 (5.0%) were diagnosed with hypertriglyceridemia. Of these, 418 (15.0%) received at least one prescription of triglyceride-lowering treatment. Over the same period, 1653 patients had at least three measurements of triglycerides ≥200 mg/dL, with only 357 (21.6%) receiving at least one prescription of triglyceride-lowering drugs in the year following the last measurement. Furthermore, 513 patients had at least one measurement of ≥500 mg/dL. Of these, only 246 (48.0%) received at least one prescription of triglyceride-lowering drugs in the year following the last measurement. In total, 3485 patients had an acute myocardial infarction (prevalence, 4.3%) in 2015. Of these, only 288 (8.3%) received at least one prescription of triglyceride-lowering drugs in the year following this event. These findings confirmed a pattern of inadequate treatment of hypertriglyceridemia in the Rome area.

Clinical trials have shown the benefits of &beta, blockers therapy in patients with heart failure reduced ejection fraction. These benefits include improved survival and a reduced need for hospitalization. Cardiac resynchronization therapy has emerged as an essential device-based therapy for symptomatic patients with heart failure reduced ejection fraction despite optimal pharmacologic treatment. The extent to which &beta, blockers are being utilized in patients receiving cardiac resynchronization therapy is not well known. In this study, we evaluate the possibility of increasing &beta, blockers doses in an unselected cohort of heart failure reduced ejection patients after cardiac resynchronization therapy capable defibrillator system implantation and the correlation between &beta, blockers treatments and clinical outcome. Methods and results: Patients with heart failure reduced ejection fraction in &beta, blockers therapy that underwent cardiac resynchronization therapy capable defibrillator system implantation between July 2008, and December 2016 were enrolled in the study. The &beta, blockers dose was determined at the time of discharge and during follow-up. Cardiovascular mortality, hospitalization for worsening heart failure or arrhythmic storm and appropriate intervention of the device, were recorded. The study cohort included 480 patients, 289 patients (60.3%) had &beta, blockers doses equal to the dose before CRT (Group 1), 191 patients (39.7%) had higher &beta, blockers doses than those before the CRT implant (Group 2). Comparing the two groups, Group 2 have lower cardiovascular mortality, heart failure-related hospitalization, and arrhythmic events than Group 1. Conclusion: After initiating CRT, &beta, blockers could be safely up-titrated at higher doses with the reduction in mortality, heart failure-related hospitalization, and arrhythmic events.

Background In hypertrophic cardiomyopathy (HCM), most of the factors associated with the risk of sudden cardiac death (SCD) are also involved in the pathophysiology of exercise limitation. The present multicentre study investigated possible ability of cardiopulmonary exercise test in improving contemporary strategies for SCD risk stratification. Methods A total of 623 consecutive outpatients with HCM, from five tertiary Italian HCM centres, were recruited and prospectively followed, between September 2007 and April 2015. The study composite end point was SCD, aborted SCD and appropriate implantable cardioverter defibrillator (ICD) interventions. Results During a median follow-up of 3.7 years (25th–75th centile: 2.2–5.1 years), 25 patients reached the end point at 5 years (3 SCD, 4 aborted SCD, 18 appropriate ICD interventions). At multivariate analysis, ventilation versus carbon dioxide relation during exercise (VE/VCO 2 slope) remains independently associated to the study end point either when challenged with the 2011 American College of Cardiology Foundation/American Heart Association guidelines-derived score (C index 0.748) or with the 2014 European Society of Cardiology guidelines-derived score (C index 0.750). A VE/VCO 2 slope cut-off value of 31 showed the best accuracy in predicting the SCD end point within the entire HCM study cohort (sensitivity 64%, specificity 72%, area under the curve 0.72). Conclusions Our data suggest that the VE/VCO 2 slope might improve SCD risk stratification, particularly in those HCM categories classified at low-intermediate SCD risk according to contemporary guidelines. There is a need for further larger studies, possibly on independent cohorts, to confirm our preliminary findings.

Background: right ventricle-pulmonary artery (RV-PA) coupling assessed by measuring the tricuspid anular plane systolic excursion (TAPSE)/pulmonary artery systolic pressure (PASP) ratio has been recently proposed as an early marker of right ventricular dysfunction in patients with heart failure with a reduced ejection fraction (HFrEF). Methods: As the effects of sacubitril/valsartan therapy on RV-PA coupling remain unknown, this study aimed to analyse the effect of this drug on TAPSE/PASP in patients with HFrEF. We retrospectively analysed all outpatients with HFrEF referred to our unit between October 2016 and July 2018. Results: At the 1-year follow-up, sacubitril/valsartan therapy was associated with a significant improvement in TAPSE (18.26 &plusmn, 3.7 vs. 19.6 &plusmn, 4.2 mm, p &lt, 0.01), PASP (38.3 &plusmn, 15.7 vs. 33.7 &plusmn, 13.6, p &lt, 0.05), and RV-PA coupling (0.57 &plusmn, 0.25 vs. 0.68 &plusmn, 0.30 p &lt, 0.01). These improvements persisted at the 2-year follow-up. In the multivariable analysis, the improvement in the RV-PA coupling was independent of the left ventricular remodelling. Conclusions: in patients with HFrEF, sacubitril/valsartan improved the RV-PA coupling, however, further trials are necessary to evaluate the role of sacubitril/valsartan in the treatment of right ventricle (RV) dysfunction either associated or not associated with left ventricular dysfunction.

Background: A blunted heart rate (HR) response is associated with an impaired peak oxygen uptake (pVO(2)), a powerful outcome predictor in hypertrophic cardiomyopathy (HCM). The present multicenter study sought to determine the prognostic role for exercise-induced HR response in HCM.Methods: A total of 681 consecutive HCM outpatients on optimized treatment were recruited. The heart failure (HF) end-pointwas death due to HF, cardiac transplantation, NYHA III-IV class progression, HF worsening leading to hospitalization and severe functional deterioration leading to septal reduction. The sudden cardiac death (SCD) end-point included SCD, aborted SCD and appropriate implantable cardioverter defibrillator discharges.Results: During a median follow-up of 4.2 years (25-75th centile: 3.9-5.2), 81 patients reached the HF and 23 the SCD end-point. Covariates with independent effects on the HF end-point were left atrial diameter, left ventricular ejection fraction, maximal left ventricular outflow tract gradient and exercise cardiac power (ECP=pVO(2)*systolic blood pressure) (C-Index= 0.807) whereas the HCM Risk-SCD score and the ECP remained associated with the SCD end-point (C-Index = 0.674). When the VO2-derived variables were not pursued, peak HR (pHR) re-entered in the multivariate HF model (C-Index= 0.777) and, marginally, in the SCD model (C-index= 0.656). A pHR= 70% of the maximum predicted resulted as the best cut-off value in predicting the HF-related events.Conclusions: The cardiopulmonary exercise test is pivotal in the HCM management, however the pHR remains a meaningful alternative parameter. A pHR < 70% identified a HCM population at high risk of HF-related events, thus calling for a reappraisal of the chronotropic incompetence threshold in HCM. (c) 2018 Elsevier B.V. All rights reserved.

Pediatric heart failure represents an important cause of morbidity and mortality in childhood. Currently, there are well-established guidelines for the management of heart failure in the adult population, but an equivalent consensus in children is lacking. In the clinical setting, ensuring an accurate diagnosis and defining etiology is essential to optimal treatment. Diuretics and angiotensin-converting enzyme inhibition are the first-line therapies, whereas beta-blockers and devices for electric therapy are less used in children than in adults. In the end-stage disease, heart transplantation is the best choice of treatment, while a left ventricular assist device can be used as a bridge to transplantation (due to the difficulties in finding organ donors), recovery (in the case of myocarditis), or destination therapy (for patients with systemic disease). Copyright (C) 2017, Taiwan Pediatric Association. Published by Elsevier Taiwan LLC.

An innovative wireless monitoring system for measuring greenhouse climatic parameters was developed to overcome the problems related to wires cabling such as presence of a dense net of wires hampering the cultivation practices, wires subjected to high temperature and relative humidity, rodents that can damage wires. The system exploits battery-powered environmental sensors, such as air temperature and relative humidity sensors, wind speed and direction, and solar radiation sensors, integrated in the contest of an 802.15.4-based wireless sensors network. Besides, a fruit diameter measurement sensor was integrated into the system. This approach guarantees flexibility, ease of deployment and low power consumption. Data collected from the greenhouse are then sent to a remote server via a general packet radio service link. The proposed solution has been implemented in a real environment. The test of the communication system showed that 0.3% of the sent data packed were lost; the climatic parameters measured with the wireless system were compared with data collected by the wired system showing a mean value of the absolute difference equal to 0.6°C for the value of the greenhouse air temperature. The wireless climate monitoring system showed a good reliability, while the sensor node batteries showed a lifetime of 530 days.

Conflict is one of the most important phenomena of social life, but it is still largely neglected by the computing community. This work proposes an approach that detects common conversational social signals (loudness, overlapping speech, etc.) and predicts the conflict level perceived by human observers in continuous, non-categorical terms. The proposed regression approach is fully Bayesian and it adopts automatic relevance determination to identify the social signals that influence most the outcome of the prediction. The experiments are performed over the SSPNet Conflict Corpus, a publicly available collection of 1,430 clips extracted from televised political debates (roughly 12 hours of material for 138 subjects in total). The results show that it is possible to achieve a correlation close to 0.8 between actual and predicted conflict perception.

Many state-of-the-art diarization systems for meeting recordings are based on the HMM/GMM framework and the combination of spectral (MFCC) and time delay of arrivals (TDOA) features. This paper presents an extensive study on how multistream diarization can be improved beyond these two sets of features. While several other features have been proven effective for speaker diarization, little efforts have been devoted to integrate them into the MFCC + TDOA state-of-the-art baseline and to the authors’ best knowledge, no positive results have been reported so far. The first contribution of this paper consists in analyzing the reasons of this, investigating through a set of oracle experiments the robustness of the HMM/GMM diarization when also other features (the modulation spectrum features and the frequency domain linear prediction features) are integrated. The second contribution of the paper consists in introducing a non-parametric multistream diarization method based on the information bottleneck (IB) approach. In contrary to the HMM/GMM which makes use of log-likelihood combination, it combines the feature streams in a normalized space of relevance variables. The previous analysis is repeated revealing that the proposed approach is more robust and can actually benefit from other sources of information beyond the conventional MFCC and TDOA features. Experiments based on the rich transcription data (heterogeneous meetings data recorded in several different rooms) show that it achieves a very competitive error of only 6.3% when four feature streams are used, compared to the 14.9% of the HMM/GMM system. Those results are analyzed in terms of error sensitivity to the stream weightings. To the authors’ best knowledge this is the first successful attempt to reduce the speaker error combining other features with the MFCC and the TDOA and the first study to show the shortcomings of the HMM/GMM in going beyond this baseline. As last contribution, the paper also addresses issues related to the computational complexity of multistream approaches.

Recently, several multi-layer perceptron (MLP)- based front-ends have been developed and used for Mandarin speech recognition, often showing significant complementary properties to conventional spectral features. Although widely used in multiple Mandarin systems, no systematic comparison of all the different approaches as well as their scalability has been proposed. The novelty of this correspondence is mainly experimental. In this work, all the MLP front-ends recently developed at multiple sites are described and compared in a systematic manner on a 100 hours setup. The study covers the two main directions along which the MLP features have evolved: the use of different input representations to the MLP and the use of more complex MLP architectures beyond the three-layer perceptron. The results are analyzed in terms of confusion matrices and the paper discusses a number of novel findings that the comparison reveals. Furthermore, the two best front-ends used in the GALE 2008 evaluation, referred as MLP1 and MLP2, are studied in a more complex LVCSR system in order to investigate their scalability in terms of the amount of training data (from 100 hours to 1600 hours) and the parametric system complexity (maximum likelihood versus discriminative training, speaker adaptative training, lattice level combination). Results on 5 hours of evaluation data from the GALE project reveal that the MLP features consistently produce improvements in the range of 15%–23% relative at the different steps of a multipass system when compared to mel-frequency cepstral coefficient (MFCC) and PLP features, suggesting that the improvements scale with the amount of data and with the complexity of the system. The integration of those features into the GALE 2008 evaluation system provide very competitive performances compared to other Mandarin systems. © 2011, IEEE.

In this paper, we describe recent work at Idiap Research Institute in the domain of multilingual speech processing and provide some insights into emerging challenges for the research community. Multilingual speech processing has been a topic of ongoing interest to the research community for many years and the field is now receiving renewed interest owing to two strong driving forces. Firstly, technical advances in speech recognition and synthesis are posing new challenges and opportunities to researchers. For example, discriminative features are seeing wide application by the speech recognition community, but additional issues arise when using such features in a multilingual setting. Another example is the apparent convergence of speech recognition and speech synthesis technologies in the form of statistical parametric methodologies. This convergence enables the investigation of new approaches to unified modelling for automatic speech recognition and text-to-speech synthesis (TTS) as well as cross-lingual speaker adaptation for TTS. The second driving force is the impetus being provided by both government and industry for technologies to help break down domestic and international language barriers, these also being barriers to the expansion of policy and commerce. Speech-to-speech and speech-to-text translation are thus emerging as key technologies at the heart of which lies multilingual speech processing.

This correspondence describes a novel system for speaker diarization of meetings recordings based on the combination of acoustic features (MFCC) and time delay of arrivals (TDOAS). The first part of the paper analyzes differences between MFCC and TDOA features which possess completely different statistical properties. When Gaussian mixture models are used, experiments reveal that the diarization system is sensitive to the different recording scenarios (i.e., meeting rooms with varying number of microphones). In the second part, a new multistream diarization system is proposed extending previous work on information theoretic diarization. Both speaker clustering and speaker realignment steps are discussed; in contrary to current systems, the proposed method avoids to perform the feature combination averaging log-likelihood scores. Experiments on meetings data reveal that the proposed approach outperforms the GMM-based system when the recording is done with varying number of microphones.

This paper aims at investigating the use of Dempster–Shafer (DS) combination rule for multi-stream automatic speech recognition. The DS combination is based on a generalization of the conventional Bayesian framework. The main motivation for this work is the similarity between the DS combination and findings of Fletcher on human speech recognition. Experiments are based on the combination of several Multi Layer Perceptron (MLP) classifiers trained on different representations of the speech signal. The TANDEM framework is adopted in order to use the MLP outputs into conventional speech recognition systems. We exhaustively investigate several methods for applying the DS combination into multi-stream ASR. Experiments are run on small and large vocabulary speech recognition tasks and aim at comparing the proposed technique with other frame-based combination rules (e.g. inverse entropy). Results reveal that the proposed method outperforms conventional combination rules in both tasks. Furthermore we verify that the performance of the combined feature stream is never inferior to the performance of the best individual feature stream. We conclude the paper discussing other applications of the DS combination and possible extensions.

Heart failure patients are predisposed to develop arrhythmias. Supraventricular arrhythmias can exacerbate the heart failure symptoms by decreasing the effective cardiac output and their control require pharmacological, electrical, or catheter-based intervention. In the setting of atrial flutter or atrial fibrillation, anticoagulation becomes paramount to prevent systemic or cerebral embolism. Patients with heart failure are also prone to develop ventricular arrhythmias that can present a challenge to the managing clinician. The management strategy depends on the type of arrhythmia, the underlying structural heart disease, the severity of heart failure, and the range from optimization of heart failure therapy to catheter ablation. Patients with heart failure, irrespective of ejection fraction are at high risk for developing sudden cardiac death, however risk stratification is a clinical challenge and requires a multiparametric evaluation for identification of patients who should undergo implantation of a cardioverter defibrillator. Finally, patients with heart failure can also develop symptomatic bradycardia, caused by sinus node dysfunction or atrio-ventricular block. The treatment of bradycardia in these patients with pacing is usually straightforward but needs some specific issue.

Speaker Diarization aims at inferring who spoke when in an audio stream and involves two simultaneous unsupervised tasks: (1) the estimation of the number of speakers, and (2) the association of speech segments to each speaker. Most of the recent efforts in the domain have addressed the problem using machine learning techniques or statistical methods (for a review see [11]) ignoring the fact that the data consists of instances of human conversations.

International audience; Ongoing major advances in plant genotyping and phenotyping lead to a better understanding of genetic architecture of agronomical traits. In this context, it is important to develop decision support tools to help breeders in implementing marker-assisted selection (MAS) projects to assemble new allele combinations. Algorithms have been developed within an interactive graphical interface to (a) trace parental QTL alleles throughout selection generations, (b) propose strategies to select the best plants based on estimated molecular scores, and (c) efficiently intermate them depending on the expected value of their progenies. By investigating multi-allelic context and diverse pedigree structure, OptiMAS enables to assemble favorable alleles issued from diverse parents and further accelerate genetic gain.

Conflict escalation in multi-party conversations refers to an increase\ud in the intensity of conflict during conversations. Here we study annotation\ud and detection of conflict escalation in broadcast political\ud debates towards a machine-mediated conflict management system.\ud In this regard, we label conflict escalation using crowd-sourced annotations\ud and predict it with automatically extracted conversational\ud and prosodic features. In particular, to annotate the conflict escalation\ud we deploy two different strategies, i.e., indirect inference and\ud direct assessment; the direct assessment method refers to a way that\ud annotators watch and compare two consecutive clips during the annotation\ud process, while the indirect inference method indicates that\ud each clip is independently annotated with respect to the level of\ud conflict then the level conflict escalation is inferred by comparing\ud annotations of two consecutive clips. Empirical results with 792\ud pairs of consecutive clips in classifying three types of conflict escalation,\ud i.e., escalation, de-escalation, and constant, show that labels\ud from direct assessment yield higher classification performance\ud (45.3% unweighted accuracy (UA)) than the one from indirect inference\ud (39.7% UA), although the annotations from both methods are\ud highly correlated (ρ = 0.74 in continuous values and 63% agreement\ud in ternary classes).

International audience; Current advances in plant genotyping lead to major progress in the knowledge of genetic architecture of traits of interest. It is increasingly important to develop decision support tools to help breeders and geneticists to conduct marker-assisted selection methods to assemble favorable alleles that are discovered. Algorithms have been implemented, within an interactive graphical interface, to 1) trace parental alleles throughout generations, 2) propose strategies to select the best plants based on estimated molecular scores, and 3) efficiently intermate them depending on the expected value of their progenies. With the possibility to consider a multi-allelic context, OptiMAS opens new prospects to assemble favorable alleles issued from diverse parents and further accelerate genetic gain.

Recognition of personality traits is a well studied problem in psychology while only recently it has been addressed by speech and language technology research. This paper describes annotation and experiments towards automatically inferring speakers personality traits in spontaneous conversations. In the first part, the work describes the annotation framework based on the Big-Five personality traits model (Extraversion, Agreeableness, Conscientiousness, Neuroticism and Openness) applied to 128 speakers from the AMI corpus. As the corpus contains rich annotations, those data can generalize previous studies based on enacted speech or dialogues. In the second part, the paper describes experiments based on various features including prosody, words n-gram, dialog acts and speech activity. Results reveal that high/low extraversion, consciousness and neuroticism traits can be automatically recognized with accuracy rate of 74.5%, 67.6% and 68.7%, respectively, while agreeableness and openness classification error rates are not statistically better than chance. Non-linguistic features (prosody, speech activity, overlaps and interruptions) outperform linguistic features (words n-gram and dialog acts) in this setup.

Automatic analysis of spoken conversations has recently searched for phenomena like agreement/disagreement in collaborative and non- conflictual discussions (e.g., meetings). This work adds a novel dimension investigating conflicts in spontaneous conversations. The study makes use of broadcasted political debates where conflicts naturally arise between participants. In the first part, an annotation scheme to rate the degree of conflict in conversations is described and applied to 12 hours of recordings. In the second part, the correlation between various prosodic/conversational features and the degree of conflict is investigated. In the third part, we perform automatic detection of the level of conflict based on those features showing an F-measure of 71.6% in three-level classification tasks.

This paper investigates the use of large TDOA feature vectors together with acoustic information in speaker diarization of meetings. TDOAs are obtained by considering all possible microphones pairs and this approach is compared with conventional TDOA features extracted w.r.t. a reference channel. The study is carried using two systems, the first based on Gaussian Mixture Modeling and the second based on the Information Bottleneck approach. Results on NIST RT06/RT07/RT09 evaluation datasets show a large speaker error reduction of 30% relative going from 14.3% to 10.8% for the first and from 12.3% to 8.2% for the second whenever the feature weighting is properly handled. Furthermore results reveal that the IB system is more robust to different number of microphones even when all pairs large TDOA vectors are used thus outperforming the HMM/GMM by 25% relative (8.2% error compared to 10.8%).

Any social interaction is characterized by roles, patterns of behavior recognized as such by the interacting participants and corresponding to shared expectations that people hold about their own behavior as well as the behavior of others. In this respect, social roles are a key aspect of social interaction because they are the basis for making reasonable guesses about human behavior. Recognizing roles is a crucial need towards understanding (possibly in an automatic way) any social exchange, whether this means to identify dominant individuals, detect conflict, assess engagement or spot conversation highlights. This work presents an investigation on language-independent automatic social role recognition in AMI meetings, spontaneous multi-party conversations, based solely on turn organization and prosodic features. At first turn-taking statistics and prosodic features are integrated into a single generative conversation model which achieves an accuracy of 59%. This model is then extended to explicitly account for dependencies (or influence) between speakers achieving an accuracy of 65%. The last contribution consists in investigating the statistical dependency between the formal and the social role that participants have; integrating the information related to the formal role in the recognition model achieves an accuracy of 68%. The paper is concluded highlighting some future directions.

MLP based front-ends have shown significant complementary properties to conventional spectral features. As part of the DARPA GALE program, different MLP features were developed for Mandarin ASR. In this paper, all the proposed frontends are compared in systematic manner and we extensively investigate the scalability of these features in terms of the amount of training data (from 100 hours to 1600 hours) and system complexity (maximum likelihood training, SAT, lattice level combination, and discriminative training). Results on 5 hours of evaluation data from the GALE project reveal that the MLP features consistently produce relative improvements in the range of 15% − 23% at the different steps of a multipass system when compared to the conventional short-term spectral based features like MFCC and PLP. The largest improvement is obtained using a hierarchical MLP approach. Index Terms: TANDEM features, Multi-Layer Perceptron, Acoustic features, GALE project, LVCSR.

Speaker diarization for meetings data are recently converging towards multistream systems. The most common complementary features used in combination with MFCC are Time Delay of Arrival (TDOA). Also other features have been proposed although, there are no reported improvements on top of MFCC+TDOA systems. In this work we investigate the combination of other feature sets along with MFCC+TDOA. We discuss issues and problems related to the weighting of four different streams proposing a solution based on a smoothed version of the speaker error. Experiments are presented on NIST RT06 meeting diarization evaluation. Results reveal that the combination of four acoustic feature streams results in a 30% relative improvement with respect to the MFCC+TDOA feature combination. To the authors’ best knowledge, this is the first successful attempt to improve the MFCC+TDOA baseline including other feature streams.

This paper investigates the detection of English spoken terms in a conversational multi-language scenario. The speech is processed using a large vocabulary continuous speech recognition system. The recognition output is represented in the form of word recognition lattices which are then used to search required terms. Due to the potential multi-lingual speech segments at the input, the spoken term detection system is combined with a module performing out-of-language detection to adjust its confidence scores. First, experimental results of spoken term detection are provided on the conversational telephone speech database distributed by NIST in 2006. Then, the system is evaluated on a multi-lingual database with and without employment of the out-of-language detection module, where we are only interested in detecting English terms (stored in the index database). Several strategies to combine these two systems in an efficient way are proposed and evaluated. Around 7% relative improvement over a stand-alone STD is achieved.

This paper investigates the use of the Variational Bayesian (VB) framework for speaker diarization of meetings data extending previous related works on Broadcast News audio. VB learning aims at maximizing a bound, known as Free Energy, on the model marginal likelihood and allows joint model learning and model selection according to the same objective function. While the BIC is valid only in the asymptotic limit, the Free Energy is always a valid bound. The paper proposes the use of Free Energy as objective function in speaker diarization. It can be used to select dynamically without any supervision or tuning, elements that typically affect the diarization performance i.e. the inferred number of speakers, the size of the GMM and the initialization. The proposed approach is compared with a conventional state-of-the-art system on the RT06 evaluation data for meeting recordings diarization and shows an improvement of 8.4% relative in terms of speaker error.

Monitoring and tracking of fleets can represent a great improvement in security and safety when transporting hazardous goods by road. In this paper we propose a complete monitoring and tracking solution for truck fleets, which is able to check at the same time the position and the mechanical status of the vehicle, as well as the conditions in the cargo bay. The system exploits battery-powered environmental sensors (temperature, humidity, pressure, gas concentration and ionising radiation levels), connected by a ZigBee-based Wireless Sensor Network. This approach guarantees flexibility, ease of deployment and low power consumption. Collected data is then sent from the vehicle to a remote server via a GPRS link. The GPS positioning system is integrated by the use of an Inertial Navigation System, which guarantees a precise estimate of the position also when the GPS signal is weak or temporarily lost. The proposed solution has been deployed in a real environment, and some performance evaluation tests have been carried out.

In this paper we propose a flexible two-level architecture for wireless communications in areas where the deployment of wired networks is not feasible, as in disaster recovery scenarios. The first layer of communication guarantees medium-range transmission, being based on the IEEE 802.11s draft for Mesh Networking. The lower one is instead optimized for short-range, low-power communication between sensor and actuator nodes, most of which are battery-powered. The system architecture and related features are described in detail, and some preliminary experimental results of the subsystems are presented.

This paper investigates the combination of two streams of acoustic features. Extending our previous work on small vocabulary task, we show that combination based on Dempster-Shafer rule outperforms several classical rules like sum, product and inverse entropy weighting even in LVCSR systems. We analyze results in terms of Frame Error Rate and Cross Entropy measures. Experimental framework uses meeting transcription task and results are provided on RT05 evaluation data. Results are consistent with what has been previously observed on smaller databases.

In this paper we aim to investigate the use of Variational Bayesian methods for audio indexing purposes. Variational Bayesian (VB) techniques are approximated techniques for fully Bayesian learning. Contrarily to non Bayesian methods (e.g. Maximum Likelihood) or partially Bayesian criterion (e.g. Maximum a Posteriori), VB benefits from important model selection properties. VB learning is based on the Free Energy optimization; Free Energy can be used at the same time as an objective function and as a model selection criterion allowing simultaneous model learning/model selection. Here we explore the use of VB learning and VB model selection in a speaker clustering task comparing results with classical learning techniques (ML and MAP) and classical model selection criteria (BIC). Experiments are run on the evaluation data set NIST-1996 HUB-4 and results show that VB can outperform classical methods.

This paper aims at comparing the Bayesian Information Criterion and the Variational Bayesian approach for scoring unknown multiple speakerclustering. Variational Bayesian learning is a very effective method that allows parameter learning and model selection at the same time. The application we consider here consists in finding the optimal clustering in a conversation where the speaker number is not a priori known. Experiments are run on synthetic data and on the evaluation data set NIST-1996 HUB-4. VB learning achieves higher score in terms of average cluster purity and average speaker purity compared to ML/BIC.

This paper aims at investigating the use of sequential clustering for speaker diarization. Conventional diarization systems are based on parametric models and agglomerative clustering. In our previous work we proposed a non-parametric method based on the agglomerative Information Bottleneck for very fast diarization. Here we consider the combination of sequential and agglomerative clustering for avoiding local maxima of the objective function and for purification. Experiments are run on the RT06 eval data. Sequential Clustering with oracle model selection can reduce the speaker error by $10\%$ w.r.t. agglomerative clustering. When the model selection is based on Normalized Mutual Information criterion, a relative improvement of $5\%$ is obtained using a combination of agglomerative and sequential clustering.

Speaker diarization of meetings recorded with Multiple Distant Microphones makes extensive use of multiple feature streams like MFCC and Time Delay of Arrivals (TDOA). Typically the combination happens using separate models for each feature stream. This work investigates if the combination of multiple feature streams can happen through the combination of multiple diarization systems performed using those features. The paper extends the previously proposed Information Bottleneck method to handle the combination of several probabilistic diarization outputs. In contrast to the conventional model-based feature combination, this technique is referred as system-based combination. Furthermore the paper introduces an hybrid model-system combination. Experiments are run on data from the Rich Transcription campaigns and show that the system based combination largely outperforms the model based combination by 37% relative. The hybrid approaches improve by 10–20%. The analysis of errors shows that the improvements come from the recordings where the individual MFCC and TDOA systems provide very different performances.

Speaker diarization of meetings can be significantly improved by overlap handling. Several previous works have explored the use of different features such as spectral, spatial and energy for overlap detection. This paper proposes a method to estimate probabilities of speech and overlap classes at a segment level which are later incorporated into an HMM/GMM baseline system. The estimation is motivated by the observation that significant portion of overlaps in spontaneous conversations take place where the amount of silence is less, e.g., during speaker changes. Experiments on the AMI corpus reveal that the probability of occurrence of overlap in a segment is inversely proportional to the amount of silence in it. Whenever this information is combined with acoustic information from MFCC features in an HMM/GMM overlap detector, improvements are verified in terms of F-measure. Furthermore the paper investigates the use of exclusion and labelling strategies based on such detector for handling overlap in diarization reporting F-measure improvements from 0.29 to 0.43 in case of exclusion and from 0.15 to 0.22 in case of labelling. Consequently speaker diarization error is reduced by 8% relative compared to the baseline based solely on acoustic information.

This paper investigates from an automatic speech recognition perspective, the most effective way of combining Multi Layer Perceptron (MLP) classifiers trained on different ranges of auditory and modulation frequencies. Two different combination schemes based on MLP are considered. The first one operates in parallel fashion and is invariant to the order in which feature streams are introduced. The second one operates in hierarchical fashion and is sensitive to the order in which feature streams are introduced. The study is carried on a Large Vocabulary Continuous Speech Recognition system for transcription of meetings data using the TANDEM approach. Results reveal that (1) the combination of MLPs trained on different ranges of auditory frequencies is more effective if performed in parallel fashion; (2) the combination of MLPs trained on different ranges of modulation frequencies is more effective if performed in hierarchical fashion moving from high to low modulations; (3) the improvement obtained from separate processing of two modulation frequency ranges (12% relative WER reduction w.r.t. the single classifier approach) is considerably larger than the improvement obtained from separate processing of two auditory frequency ranges (4% relative WER reduction w.r.t. the single classifier approach). Similar results are also verified on other LVCSR systems and on other languages. Furthermore, the paper extends the discussion to the combination of classifiers trained on separate auditory-modulation frequency channels showing that previous conclusions hold also in this scenario.

This paper investigates the combination of evidence coming from different frequency channels obtained filtering the speech signal at different auditory and modulation frequencies. In our previous work \cite{icassp2008}, we showed that combination of classifiers trained on different ranges of {\it modulation} frequencies is more effective if performed in sequential (hierarchical) fashion. In this work we verify that combination of classifiers trained on different ranges of {\it auditory} frequencies is more effective if performed in parallel fashion. Furthermore we propose an architecture based on neural networks for combining evidence coming from different auditory-modulation frequency sub-bands that takes advantages of previous findings. This reduces the final WER by 6.2\% (from 45.8\% to 39.6\%) w.r.t the single classifier approach in a LVCSR task.

Spoken cultural heritage can present considerably heterogeneous content as tales, stories, recitals, poems, theatrical representations and other form of folk literature. This work investigates the automatic detection and classification of those data type in large spoken audio archives. The corpus used for this study consists of 90 radio broadcast shows collected for preserving a large variety of Swiss French dialects. Given the variability of the language spoken in the recordings, the paper proposes a language-independent system based on structural features obtained using a speaker diarization system and various acoustic/prosodic features. Results reveal that such a system can achieve an F-measure equal to 0.85 (Precision 0.88/Recall 0.84) in retrieving folk literature in those archives. Prosodic features appear more effective and complementary to structural features. Furthermore, the paper investigates whether the same approach can be used to label speech segments into five large classes (Storytelling, Poetry, Theatre, Interviews, Functionals) showing F-measures ranging from 0.52 to 0.88. As last contribution, prosodic features for disambiguating between spoken prose and spoken poetry are investigated. In summary the study shows that simple structural and acoustic/prosodic features can be used to effectively retrieve and label folk literature in broadcast archives.

This work aims at investigating the automatic recognition of speaker role in meeting conversations from the AMI corpus. Two types of roles are considered: formal roles, fixed over the meeting duration and recognized at recording level, and social roles related to the way participants interact between themselves, recognized at speaker turn level. Various structural, lexical and prosodic features as well as Dialog Act tags are exhaustively investigated and combined for this purpose. Results reveal an accuracy of 74% in recognizing the speakers formal roles and an accuracy of 66% (percentage of time) in correctly labeling the social roles. Feature analysis reveals that lexical features provide the higher performances in formal/functional role recognition while prosodic features provide the higher performances in social role recognition. Furthermore results reveal that social role recognition in case of rare roles in the corpus can be improved through the use of lexical and Dialog Act information combined over short time windows.

This paper investigates the automatic detection of English spoken terms in a multi-language scenario over real lecture recordings. Spoken Term Detection (STD) is based on an LVCSR where the output is represented in the form of word lattices. The lattices are then used to search the required terms. Processed lectures are mainly composed of English, French and Italian recordings where the language can also change within one recording. Therefore, the English STD system uses an Out-Of-Language (OOL) detection module to filter out non-English input segments. OOL detection is evaluated w.r.t. various confidence measures estimated from word lattices. Experimental studies of OOL detection followed by English STD are performed on several hours of multilingual recordings. Significant improvement of OOL+STD over a stand-alone STD system is achieved (relatively more than 50% in EER). Finally, an additional modality (text slides in the form of PowerPoint presentations) is exploited to improve STD.

This paper investigates detection of English keywords in a conversational scenario using a combination of acoustic and LVCSR based keyword spotting systems. Acoustic KWS systems search predefined words in parameterized spoken data. Corresponding confidences are represented by likelihood ratios given the keyword models and a background model. First, due to the especially high number of false-alarms, the acoustic KWS system is augmented with confidence measures estimated from corresponding LVCSR lattices. Then, various strategies to combine scores estimated by the acoustic and several LVCSR based KWS systems are explored. We show that a linear regression based combination significantly outperforms other (model-based) techniques. Due to that, the relative number of false-alarms of the combined KWS system decreased by more than 50% compared to the acoustic KWS system. Finally, an attention is also paid to the complexities of the KWS systems enabling them to potentially be exploited in real-detection tasks.