scholarly journals Mutual impact of acoustic and linguistic representations for continuous emotion recognition in call-center conversations

2021 ◽  
Author(s):  
Marie Tahon ◽  
Manon Macary ◽  
yannick Estève ◽  
Daniel Luzzati
Keyword(s):  
Call Center ◽  
Real Life ◽  
Biological Response ◽  
Textual Information ◽  
Unseen Data ◽  
Linguistic Representations ◽  
High Contribution ◽  
Customer Services ◽  
Audio Information ◽  
Linguistic Content

<div> <div> <div> <p>The goal of our research is to automaticaly retrieve the satisfaction and the frustration in real-life call-center conversations. This study focuses an industrial application in which the customer satisfaction is continuously tracked down to improve customer services. To compensate the lack of large annotated emotional databases, we explore the use of pre-trained speech representations as a form of transfer learning towards AlloSat corpus. Moreover, several studies have pointed out that emotion can be detected not only in speech but also in facial trait, in biological response or in textual information. In the context of telephone conversations, we can break down the audio information into acoustic and linguistic by using the speech signal and its transcription. Our experiments confirms the large gain in performance obtained with the use of pre-trained features. Surprisingly, we found that the linguistic content is clearly the major contributor for the prediction of satisfaction and best generalizes to unseen data. Our experiments conclude to the definitive advantage of using CamemBERT representations, however the benefit of the fusion of acoustic and linguistic modalities is not as obvious. With models learnt on individual annotations, we found that fusion approaches are more robust to the subjectivity of the annotation task. This study also tackles the problem of performances variability and intends to estimate this variability from different views: weights initialization, confidence intervals and annotation subjectivity. A deep analysis on the linguistic content investigates interpretable factors able to explain the high contribution of the linguistic modality for this task. </p> </div> </div> </div>

scholarly journals Mutual impact of acoustic and linguistic representations for continuous emotion recognition in call-center conversations

2021 ◽  
Author(s):  
Marie Tahon ◽  
Manon Macary ◽  
yannick Estève ◽  
Daniel Luzzati
Keyword(s):  
Call Center ◽  
Real Life ◽  
Biological Response ◽  
Textual Information ◽  
Unseen Data ◽  
Linguistic Representations ◽  
High Contribution ◽  
Customer Services ◽  
Audio Information ◽  
Linguistic Content

<div> <div> <div> <p>The goal of our research is to automaticaly retrieve the satisfaction and the frustration in real-life call-center conversations. This study focuses an industrial application in which the customer satisfaction is continuously tracked down to improve customer services. To compensate the lack of large annotated emotional databases, we explore the use of pre-trained speech representations as a form of transfer learning towards AlloSat corpus. Moreover, several studies have pointed out that emotion can be detected not only in speech but also in facial trait, in biological response or in textual information. In the context of telephone conversations, we can break down the audio information into acoustic and linguistic by using the speech signal and its transcription. Our experiments confirms the large gain in performance obtained with the use of pre-trained features. Surprisingly, we found that the linguistic content is clearly the major contributor for the prediction of satisfaction and best generalizes to unseen data. Our experiments conclude to the definitive advantage of using CamemBERT representations, however the benefit of the fusion of acoustic and linguistic modalities is not as obvious. With models learnt on individual annotations, we found that fusion approaches are more robust to the subjectivity of the annotation task. This study also tackles the problem of performances variability and intends to estimate this variability from different views: weights initialization, confidence intervals and annotation subjectivity. A deep analysis on the linguistic content investigates interpretable factors able to explain the high contribution of the linguistic modality for this task. </p> </div> </div> </div>

scholarly journals The Effect of Irrelevant Environmental Noise on the Performance of Visual-to-Auditory Sensory Substitution Devices Used by Blind Adults

2019 ◽  
Vol 32 (2) ◽  
pp. 87-109 ◽  
Author(s):  
Galit Buchs ◽  
Benedetta Heimler ◽  
Amir Amedi
Keyword(s):  
Background Noise ◽  
Visual Information ◽  
Real Life ◽  
Sensory Substitution ◽  
Auditory Modality ◽  
Auditory Information ◽  
Visual Rehabilitation ◽  
Actual Use ◽  
Significant Performance ◽  
Audio Information

Abstract Visual-to-auditory Sensory Substitution Devices (SSDs) are a family of non-invasive devices for visual rehabilitation aiming at conveying whole-scene visual information through the intact auditory modality. Although proven effective in lab environments, the use of SSDs has yet to be systematically tested in real-life situations. To start filling this gap, in the present work we tested the ability of expert SSD users to filter out irrelevant background noise while focusing on the relevant audio information. Specifically, nine blind expert users of the EyeMusic visual-to-auditory SSD performed a series of identification tasks via SSDs (i.e., shape, color, and conjunction of the two features). Their performance was compared in two separate conditions: silent baseline, and with irrelevant background sounds from real-life situations, using the same stimuli in a pseudo-random balanced design. Although the participants described the background noise as disturbing, no significant performance differences emerged between the two conditions (i.e., noisy; silent) for any of the tasks. In the conjunction task (shape and color) we found a non-significant trend for a disturbing effect of the background noise on performance. These findings suggest that visual-to-auditory SSDs can indeed be successfully used in noisy environments and that users can still focus on relevant auditory information while inhibiting irrelevant sounds. Our findings take a step towards the actual use of SSDs in real-life situations while potentially impacting rehabilitation of sensory deprived individuals.

scholarly journals Dark Side of Corporate in Chetan Bhagat’s One Night @ Call Centre

2019 ◽  
Vol 8 (1) ◽  
pp. 58-60
Author(s):  
R Ranjith kumar ◽  
B Vasanthakumar
Keyword(s):  
Life Style ◽  
Call Center ◽  
Real Life ◽  
Dark Side ◽  
Call Centre ◽  
The Novel ◽  
Comfort Zone ◽  
Night Call ◽  
Set Up ◽  
Night Shifts

The purpose of this article is to show case the corporate life style with the living ambience in society with the perspectives of social, economical, legal, and political and so on and so forth. The story was about six people working in a call center. The present paper focuses on the living style of people especially in a corporate world with the characters such as Vroom, Isha, Radhika, Syam, Bakshi and Priyanka. The setting of the novel One Night @ Call Centre is a resemblance of the posh culture of the 21st century. All these characters make the readers to be in a comfort zone by showing the contemporary issues such as work stress, love, night shifts, friendship, In-laws restrictions, Luxurious life style, Craze for fanciful life. In a nutshell it is a story of almost lost love, thwarted ambitions, negligence of family affection, stress of a patriarchal set up, an insight on the lifestyle of youth of this country and the work ambience of a globalized office. Chetan Bhagat succeeds in representing such an advanced as well as corporate style of life through all these characters by narrating the real life situations among this group of people practically. The novel also stresses on the love affairs as well as increasing rate of the divorce as well as break ups in love which are become a common issues now a days. There are certain other aspects where the writer showcases the elements of exploitation of the educated employees in call centers in this novel of One Night @ Call Centre, which are mainly highlighted in this article with few relevant examples.

scholarly journals Hybrid neural network based on novel audio feature for vehicle type identification

2021 ◽  
Vol 11 (1) ◽  
Author(s):  
Haoze Chen ◽  
Zhijie Zhang
Keyword(s):  
Neural Networks ◽  
Visual Information ◽  
Recognition Rate ◽  
Real Life ◽  
Audio Signal ◽  
Training Data ◽  
Identification System ◽  
The Novel ◽  
Vehicle Information ◽  
Audio Information

AbstractDue to the audio information of different types of vehicle models are distinct, the vehicle information can be identified by the audio signal of vehicle accurately. In real life, in order to determine the type of vehicle, we do not need to obtain the visual information of vehicles and just need to obtain the audio information. In this paper, we extract and stitching different features from different aspects: Mel frequency cepstrum coefficients in perceptual characteristics, pitch class profile in psychoacoustic characteristics and short-term energy in acoustic characteristics. In addition, we improve the neural networks classifier by fusing the LSTM unit into the convolutional neural networks. At last, we put the novel feature to the hybrid neural networks to recognize different vehicles. The results suggest the novel feature we proposed in this paper can increase the recognition rate by 7%; destroying the training data randomly by superimposing different kinds of noise can improve the anti-noise ability in our identification system; and LSTM has great advantages in modeling time series, adding LSTM to the networks can improve the recognition rate of 3.39%.

scholarly journals A General Framework to Compare Announcement Accuracy: Static vs. LES-Based Announcement

2020 ◽  
Author(s):  
Achal Bassamboo ◽  
Rouba Ibrahim
Keyword(s):  
Operations Management ◽  
Call Center ◽  
Service Providers ◽  
Waiting Times ◽  
Real Life ◽  
Weighted Average ◽  
Superior Performance ◽  
Service Discipline ◽  
Priority Service ◽  
Real Time Information

Service providers often share delay information, in the form of delay announcements, with their customers. In practice, simple delay announcements, such as average waiting times or a weighted average of previously delayed customers, are often used. Our goal in this paper is to gain insight into when such announcements perform well. Specifically, we compare the accuracies of two announcements: (i) a static announcement that does not exploit real-time information about the state of the system and (ii) a dynamic announcement, specifically the last-to-enter-service (LES) announcement, which equals the delay of the last customer to have entered service at the time of the announcement. We propose a novel correlation-based approach that is theoretically appealing because it allows for a comparison of the accuracies of announcements across different queueing models, including multiclass models with a priority service discipline. It is also practically useful because estimating correlations is much easier than fitting an entire queueing model. Using a combination of queueing-theoretic analysis, real-life data analysis, and simulation, we analyze the performance of static and dynamic announcements and derive an appropriate weighted average of the two which we demonstrate has a superior performance using both simulation and data from a call center. This paper was accepted by Vishal Gaur, operations management.

scholarly journals The Relevance of Pedagogical Translation for the Development of Bilingual Education in Costa Rica

2018 ◽  
Vol 23 (1) ◽  
Author(s):  
Ruth Cristina Hernández-Ching
Keyword(s):  
Second Language ◽  
Costa Rica ◽  
Private Colleges ◽  
Call Center ◽  
Real Life ◽  
Costa Rican ◽  
Effective Technique ◽  
Public And Private ◽  
Public School Systems ◽  
Short Period

The article reflects on bilingualism in Costa Rica in recent years in light of the latest versions of the Reports on the Costa Rican Public School Systems (2011, 2013, 2015 y 2017). Successful contributions of several national and international researches, where teaching translation as effective technique for developing communication skills is proposed, are discussed. Also, the article reviews major historical landmarks of translation in second language teaching. There are programs in the Ministry of Public Education (MEP), public and private colleges, schools and universities, but there is a tendency to associate the use of translation in teaching only with the grammatical method. Later studies could be oriented to compare the progress between populations that have acquired the language as a second language and have worked for a short period of time in a call center, in tourism, or in real life activities where they have to translate or interpret in real mode, compared to those that do not.

The British paid labour market1

2019 ◽  
pp. 45-62
Author(s):  
Eva A. Duda-Mikulin
Keyword(s):  
Labour Force ◽  
Qualitative Interviews ◽  
Real Life ◽  
Foreign Born ◽  
Women Migrants ◽  
Economic Migration ◽  
Customer Services ◽  
The Uk ◽  
The Right ◽  
The Eu

Chapter three explores the British paid labour market and more specifically economic migration to the UK and its impact with the message that migrants contribute through taxation and alleviating labour and skills shortages. I discuss existing statistical data on UK’s labour force and its characteristics. This quantitative data is complemented with rich qualitative accounts from recent Polish women migrants to the UK. Different sectors of the economy are explored, in particular agriculture, hospitality, customer services and healthcare. These are said to be most reliant on workforce from the EU. Data on population characteristics is analysed taking into account the fact that it is ageing rapidly as is the rest of Europe. This increases the need for foreign-born labour to take on jobs unpopular with British workers, particularly when the EU labour force is younger and fitter in comparison to UK-born workers. This also suggests that after Brexit the UK is likely to experience issues with staff recruitment and labour shortages in certain areas of the economy. The chapter is supported by extracts from qualitative interviews with women migrants from Poland with the aim to bring in real-life stories from those who took advantage of the right to free movement.

scholarly journals P399 Endoscopic and histologic outcome in tofacitinib treated refractory moderate-to-severe ulcerative colitis: A prospective real-life cohort

2020 ◽  
Vol 14 (Supplement_1) ◽  
pp. S369-S370
Author(s):  
B Verstockt ◽  
A Outtier ◽  
J Lefrère ◽  
J Sabino ◽  
S Vermeire ◽  
...  
Keyword(s):  
Ulcerative Colitis ◽  
Kinase Inhibitor ◽  
Remission Rate ◽  
Real Life ◽  
Biological Response ◽  
Janus Kinase ◽  
High Dose ◽  
Faecal Calprotectin ◽  
Severe Ulcerative Colitis ◽  
Alternative Treatment Strategy

Abstract Background The pan-Janus kinase inhibitor tofacitinib (TFC) has recently been approved for treatment moderate-to-severe ulcerative colitis (UC). Real-life data are limited, especially on endoscopic and histologic outcome. We report the efficacy and safety of TFC in refractory UC patients, and assessed potential clinical predictors of response. Methods Thirty-five UC patients, all refractory to anti-TNF and vedolizumab, were prospectively included (Table 1). All received TFC 10mg BID till week 8, and were endoscopically assessed at baseline (Mayo endoscopic sub-score ≥2) and week 8. Biological response was defined as a 50% decrease in faecal calprotectin (fCal) or fCal &lt;250 mg/g, and biological remission as a fCal &lt;250 mg/g at week 8. The endoscopic response was defined as Mayo endoscopic sub-score of ≤1, endoscopic remission as a sub-score of 0. Histologic remission was defined as a numeric Geboes score ≤6 (similar to ≤2A.0). A non-response imputation and last observation carried forward analysis was applied. Results The Mayo endoscopic sub-score decreased significantly by week 8 (p = 0.004), resulting in an endoscopic response and remission rate of 22.9% and 17.2% respectively. Histological remission was seen in 14.8% of patients. Faecal calprotectin decreased from 1386 mg/g down to 568 mg/g by week 4 (p = 0.03) (Table 2), but not further down by week 8 (703 mg/g, p = 0.5) with a biological response and remission rate of 52.9% and 38.2%. Half of the patients with a PNR to one anti-TNF (10 out of 20) did discontinue TFC because of PNR. However, PNR to two anti-TNF agents almost exclusively (4 out of 5) resulted in PNR to TFC. In contrast, only 3 out of 8 vedolizumab PNR experienced PNR to TFC. Ultimately, 48.6% of all included patients discontinued TFC therapy after a median of 15.9 [12.4–26.6] weeks, all but one due to PNR, of whom 9/17 (52.9%) required colectomy. In multivariate analysis, a higher baseline albumin and a lower Mayo endoscopic sub-score were independent predictors of endoscopic (OR 1.06, p = 0.02; OR 0.59, p = 0.003) and biological remission (OR 1.06, p = 0.03; OR 0.57, p = 0.01). By week 8, creatinine kinase significantly increased (p = 0.001), whereas the lipid profile was not significantly affected. One patient suffered from vaginal herpes infection, and one patient treated with TFC and high dose steroids developed disseminated nocardia, pneumocystic jiroveci, cutaneous zoster and varicella pneumoniae. Conclusion TFC can induce biologic, endoscopic and histologic remission in refractory UC, though clinical and predictive molecular markers are required to identify the right patients. In patients with prior PNR to 2 anti-TNF agents, TFC does not seem an alternative treatment strategy.

scholarly journals Performance Evaluation and Dimensioning of Systems through Kernel Estimation

2011 ◽  
Vol 2011 ◽  
pp. 1-20 ◽  
Author(s):  
G. M. Gontijo ◽  
G. S. Atuncar ◽  
F. R. B. Cruz ◽  
L. Kerbache
Keyword(s):  
Performance Measures ◽  
Call Center ◽  
Kernel Smoothing ◽  
Blocking Probability ◽  
Queueing Systems ◽  
Real Life ◽  
Kernel Estimation ◽  
Queue Size ◽  
Number Of Customers ◽  
Arrival Pattern

We extend the analysis of queueing systems for real-life situations, where the arrival pattern of customers is unknown. In real systems, we must understand how the choice of a method of estimation influences the configuration of the system. Using kernel smoothing, we evaluate algorithms to estimate performance measures of a system, including the invariant probability distribution of the number of customers in the system, the blocking probability, the average queue size, and the average client queue time. We successfully apply the method to the arrivals to a call center to plan and improve the performance of these important queueing systems.

Sign in / Sign up

Export Citation Format

Share Document