George C. Alexandropoulos Single Presenter

The increasingly demanding objectives for sixth Generation (6G) wireless communication networks have spurred recent research activities on novel wireless hardware architectures and wireless connectivity paradigms. Among them belong the Reconfigurable Intelligent Surfaces (RISs), which are artificial planar structures with integrated electronic circuits that can be programmed to manipulate an incoming electromagnetic field in a wide variety of functionalities. Incorporating RISs in wireless networks has been recently advocated as a revolutionary means to transform any naturally passive wireless communication environment to an active one. This can be accomplished by deploying cost-effective and easy to coat RISs to the environment\\\'s 3D components (e.g., building facades and room ceilings), thus, offering increased environmental intelligence for the scope of diverse wireless networking objectives. In this talk, the potential of RISs for enabling smart wireless connectivity schemes for IoT applications will be presented, emphasizing on their coverage extension capability for massive low-power devices and their perspective for enabling ubiquitous sensing in conjunction with communications. The various available hardware designs for RISs, their available modeling approaches, and their implications in the design of communication and sensing algorithms will be discussed. In addition, representative signal processing approaches for efficient RIS configurations will be presented together with candidate network architectures including RISs. Future directions with RIS-empowered IoT applications for 6G will be highlighted.

Karl Stolley Single Presenter

This talk will provide a diagnostic demonstration of the finalized WebRTC 1.0 specification as it is currently implemented in leading desktop and mobile browsers. The talk will also outline the state of WebRTC-adjacent API implementations, such as MediaStreams and Blobs. Attendees will receive web-available example code illustrating the remaining necessary polyfills and fallbacks for handling perfect negotiation, binary data channels, and media streams in older browsers as well as those with lagging implementations, such as Safari on both MacOS and iOS. The talk will conclude with a brief look at open issues and WebRTC bugs on file with major browsers, and project their likely resolution timelines.

Tim Panton Single Presenter

Not all webRTC usage is server based. This talk describes some niche decentralized webRTC projects:A baby monitor that still works when your ISP connection goes down. An SMS triggered video call webapp. A webapp for Yoga Teachers.I\\\'ll show how each of these web apps utilises specific webRTC features (e.g ICE, self signed certificates, bandwidth estimation, web audio integration etc) to provide a low cost, high quality service that is easy to use, secure, private and permissionless.These edge based projects also take advantage of increasing uplink bandwidth, with users to runnning their own services (in the browser or on a dedicated device), allowing them to choose their own business model rather than having to monetise via ads and be beholden to the increasing gamification of the major streaming players.I\\\'ll demo a couple of the apps and dig into the technical detail of how they work

Dean Bubley Single Presenter

Despite the rise of video, voice communications isn’t dead – but it is changing rapidly, adopting more useful characteristics than a plain-vanilla phone call. There are three types of “context-based RTC” that are emerging:Voice *in* a context – for example, voice-chat embedded inside a game, realtime drop-in voice chat like Clubhouse and Twitter Spaces, or push-to-talk voice for a private 4G network at a port or industrial complex. Voice *using* contextual data & IoT – for instance a personal safety-alarm linked to a fall sensor, or a contact centre using emotional analytics to better deal with customer complaints Voice *optimised* for a context – for instance, adjusting the noise-cancellation setting to the user’s ambient surroundingsTogether, context makes voice communications more useful, productive and fun. It also works better than video for many use-cases

Anthony C.K. Soong Single Presenter

The deployment of 5G is occurring as we are undergoing the fourth industrial revolution. This industrial revolution pertains to the merging of the industrial physical world with the biological and the digital worlds. One characteristic of which is already starting to emerge, digitalization and artificial intelligence have evolved to a point where we can control the physical world from virtual reality interfaces. In the not too distance future, our virtualized world will be able to also influence biological entities. In the future, we envision a world where our senses are augmented by different virtual reality universes. For example, scientists and engineers are already working to realize the fantastical world, the so called metaverse, of the film Ready Player One. One of the contributing research and development thrust for realizing this fantastical world is time sensitive internet of things (IoT). This keynote will discuss the need of time sensitive IoT for metaverse and the support of which in the current 5G system. It will not only dive into how URLLC and IoT are supported in the physical layer but will look at the support of time sensitive IoT in the wireless system from an end-to-end perspective including the support of time sensitive networks by the 5G core. The talk will conclude with challenges that are still facing the industry in this area where deep research is still required to develop the key solutions.

Paul-Louis Ageneau Single Presenter

Peer-to-Peer connections between browser-based clients, and optionally native clients, is a specific usage of WebRTC. Different solutions, like WebTorrent, have emerged to leverage WebRTC Data Channels for this purpose.However, a signaling server, also called "tracker" in such a Peer-to-Peer setup, is still necessary, which creates a central single point of failure in an otherwise distributed system.In this presentation, we will explore the challenges in designing a WebRTC Peer-to-Peer network without a central signaling server, or "trackerless". We will present Legio, an experimental distributed signaling protocol, and discuss the limitations of such an unconventional architecture.

Amisha Agarwal Single Presenter

Applications for Internet of Things (IoT) ecosystems are everywhere—agriculture, manufacturing, healthcare, retail, supply chain, transportation and many more. Speed and efficiency are paramount in such a context, not only to process but also to act on IoT data.Traditionally, IoT-like systems leverage propositional logic from various components to act on the data received. While this approach is effective, it can take a great deal of time and engineering effort to develop systems that can act on such data.We will discuss how to leverage state-of-the-art data science components using REFIT’s architecture to consume streaming data from ingestion, to training and inference, and visualization of AI models. REFIT enables seamless integration of the data science darling python and computationally efficient Scala. AI processes such as data engineering need to be written only once in python and then they are propagated to training and inference.REFIT allows data injection through low latency streaming; the streamed data is stored in a high throughput persistent storage database which is augmented with other static data sources to train predictive AI models. The current trained model is pushed to the main processing unit that conducts predictions on real time data. The various data streams and predictions are stored in a persistent storage and can be visualized in real time in a dashboard.REFIT’s data streaming architecture ensures the seamless deployment of new predictive models to the stream processing pipeline. We present in detail the use cases of monitoring and predicting traffic congestion and estimating fine-granular demand for bikes of a bike-sharing company.

Alan Quayle Single Presenter

Running a global hackathon through a pandemic is not easy. Actually, it's really hard. Especially given all the uncertainty, disparate social distancing requirements, and infection / vaccination rates around the world. But enough of my grousing ;) TADHack Global 2021 ran the weekend before Enterprise Connect; which gave us a chance to show-off the global hacking talent to the programmable communications industry. We'll review some of the impressive creations, that are making a difference in people's lives.

Dieff Vital Moderator
Hamza Soury Panelist
Mahmoud Wagih Panelist
Shuhui Grace Yang Panelist

Internet of things (IoT) has been a topic of great interest for the past decade. It enables the connectivity between objects (devices) to improve our daily life. According to Al-Fuqaha et al. (2015), it is the combination of identification (RFID), sensing, communication, computation, services, and semantics. The identification permits the exact naming of the objects to match services with their demand. The sensing layer allows the collection of data to be stored at a warehouse, database or cloud for analysis and decision-making. The communication layer facilitates the connection between the objects to deliver the smart services. The computation layer is to process the data gathered from the objects via sensing. The service layer is to cater to specific demand. Finally, the semantics are referred to the ability to extract knowledge intelligently from different objects to provide the required services. All these layers are the attributes of the idea aiming to improve our life by connecting devices, technologies, and applications. Various research topics have been studied in academia and at the industrial level to develop hardware/devices and technologies to bring the idea to life. This proposal aims at organizing a panel of academic and industrial researchers on various IoT topics who will talk about the state of the IoT knowledge where they will take us to the research journey where it all started and the progress that has been made to date. This panel will feature three speakers who are Shuhui Grace Yang, professor of computer science at Purdue University Northwest, Hamza Soury, Senior Electronic Engineer Integrated Components and Solutions Division Caterpillar Inc., and Mahmoud Wagih, Research Fellow from the University of Southampton, UK. They will give an industrial and academic perspectives on the research that has been undertaken on IoT. The panel will be led by Dieff Vital, Bridge-to-Faculty Postdoc at the University of Illinois Chicago.

Lorenzo Mangani Single Presenter

HOMER provides an open-source capture lego set designed to build open and distributed monitoring, troubleshooting and data mining integrations for modern cloud Voice, VoIP, Mobile and RTC architectures, providing everyone with the same tools used by some of the largest enterprises and operators worldwide. Let\\\'s discover and learn how to use the QXIP stack components and their many integrations to build your dream end-to-end monitoring stack!

Justin Williams Single Presenter

CPaaS generally provide comprehensive functionality via high level APIs. This provides a good abstraction of the inner workings of WebRTC away from application code. This then allows application developers to focus on the use cases and business logic of the application.While CPaaS libraries provide high enablement of building WebRTC applications, maintaining that application code can become difficult without adhering to general software best practices. CPaaS libraries generally provide a service like structure of APIs which may be different than what many frontend frameworks generally optimize for.This presentation will look at design patterns, best practices, and philosophies for building maintainable WebRTC applications.

Bill Mertka Moderator
Chuck Vick Panelist
Eric Hagerson Panelist
Fabricio Velez Panelist

This panel of distinguished experts will discuss the different Next Generation 9-1-1 Core Services (NGCS) \\\"standards families\\\" that exist in the industry today, the need for end-to-end standards to facilitate the integration of true NG9-1-1 from OSPs all the way to first responders, the state of the technical art in field, and, some of the policy, regulatory, and operational implications of such a technical \\\"state of affairs\\\" and what all of this might mean for the industry in the light of recent legislative, policy, regulatory, and technical activity in NG9-1-1 as whole.

Brooks Shannon Single Presenter

During this session, learn how PSAP virtualization enabled by cloud native technology helps PSAPs maintain daily operations regardless of circumstances. The presentation will cover the challenges that an outage causes in the 9-1-1 call flow journey – when dialing, answering, and dispatching a response to the 9-1-1 call – and how to overcome them, including how PSAPs can get notified about the intent to make an emergency call even when a 9-1-1 network is inoperable. Attendees will hear lessons learned from multiple states and regions across the country that have employed cloud technology, accurate location, and authoritative GIS data to ensure continuity of their operations.

Lorenzo Miniero Single Presenter

Several different technologies have been designed over the years to implement broadcasting of streams to large audiences over the Internet. RTMP, HLS, DASH and others have been the most commonly used in recent times. That said, the increasing need for closer-to-realtime streaming solutions that minimize delay as much as possible have recently pushed for a change, and while WebRTC as technology was initially conceived for peer-to-peer conversations, the pieces are all there to use it as a broadcasting tool as well.This presentation will describe the problem space and challenges, and introduce WHIP (WebRTC-HTTP Ingest Protocol), a protocol proposal to partially address it. It will then address the related standardization efforts in the WISH (WebRTC Ingest Signaling over HTTPS) IETF working group, and discuss a sample implementation of the protocol on top of Janus, an open source WebRTC server.

Bonny Angell Panelist
Brooks Shannon Single Presenter
Carol Davids Moderator
Jeanna Green Panelist
Joseph Cusimano Panelist
Wolfgang Kampichler Panelist

End to end testing of Next Generation 911 networks is an essential part of the current effort to port 911 and 112 type services off of circuit switched and onto IP-based networks. The goal is to ensure that all the elements - no matter which vendors produced them and which service providers\\\' networks they traverse - cooperate to carry the call and the administrative, personal and medical information that internet protocol and web technologies enable. Our speakers will describe US-based and trans-Atlantic Next Generation emergency call testing performed this year in the RTC Lab on the lab\\\'s NG911 test bed. These tests explored interesting scenarios that will likely occur as people roam the country and the world while using phone services based in their home states or countries. During the demonstration portion, you will see calls from a T-Mobile cell tower that are routed to appropriate PSAPs associated with the RTC LAb\\\'s testbed.

Dan Jenkins Single Presenter

Real Time Media is more than just delivering media from A to B using codec C. In this session I'll go over some of the tricks of the trade in how we tackle offering "professional grade" video and audio to users of Broadcaster.VC and how you can go about applying some of these tricks to your app.

Varun singh Single Presenter

18 months into the pandemic. These are Lessons learnt and improvements talk. We will cover — How we enabled products to be more reliable, set better expectation with endusers, close call quality tickets with high customer satisfaction with a small team.

Elizabeth English Panelist
Mark Fletcher, ENP Panelist
Martha Buyer Panelist
Roberta Fox Panelist

9-1-1 Expert panelists will provide perspective from their years of practical experience. What is it, what can the next generation technologies do, and how does it apply to me and my organization?This presentation will provide attendees with understanding from the Public Safety, Regulatory, and Enterprise perspective. Understand the pieces and how they fit together from roadmap, design, planning, procurement and implementation through day 2 operations.

Sean DuBois Single Presenter

WebRTC is the lingua franca of conferencing. Almost every new project coming out uses it, but what else can you do with WebRTC? Learn about the interesting projects on the horizon. Pion is being used to power VPNs, cloud dev enviroments, game streaming, robots and massively scaled Kubernetes WebRTC installs. We will talk about these interesting projects and how Pion\\\'s simple composable API could power yours next.

Bernard Aboba Panelist
Chris Cunningham Panelist
Paul Adenot Panelist

For decades, streaming and realtime communications applications have utilized distinct protocol architectures and APIs. However, with the pandemic condensing a decade of innovation into a few months, the worlds of streaming and realtime communications have converged as \\\"low latency\\\" streaming applications have proliferated.The merging of realtime and streaming technologies has exposed the need for a common set of tools to allow developers to build next generation communications applications.This talk will introduce one of the central tools in the new web media toolbox, WebCodecs. The WebCodecs API provides low-level access to the codecs already available in browsers. When combined with other APIs under development (such as \\\"Breakout Box\\\", WebTransport and WebGL) and existing APIs such as Data Channel, Media Capture and Canvas, WebCodecs can be used to improve existing applications as well as to create new applications blending streaming and realtime communications techniques.We will cover the WebCodecs API as well as introducing other APIs often used with it. The session will include demos illustrating what the new APIs are capable of.

Bill Mertka Single Presenter

The rollout of NG9-1-1 systems based upon NENA\\\'s i3 standard, and associated standards from various other SDOs, has helped to move public safety out of the obsolescent single-media, bandwidth-limited legacy world, toward an IP-enabled multimedia world, but this only goes part-way towards fulfilling the NG "vision." What's needed is additional application of modern state-of-the-art technologies like Artificial Intelligence (AI) and Machine Learning (ML) to address the "deluge" of new multimedia information that will be processed by the Emergency Communications Center (ECC, a new name and concept for today\\\'s PSAPs), most of which in coming years will be coming from Internet of Things (IoT) devices and other non-human initiated sources. A new era is coming for the ECC, and Public Safety officials can no long look at procuring legacy, standalone applications like Call Handling and CAD, but need instead to start thinking in terms of an integrated "ECC Operating System" that takes full advantage of the new information enabled by NG9-1-1 and processes it into the actionable intelligence in the field needed to improve public safety outcomes for all stakeholders. This presentation / course will introduce some of the new concepts, where they fit in a "i4 system," so to speak and how new technologies will drive new public safety workflow, all aimed at the next step "beyond i3" and enabling the fulfillment of the NG public safety vision.

Hender Jimenez Single Presenter

Television white spaces (TVWS) refers to frequencies in the very high frequency (VHF) and ultra-high frequency (UHF) television bands not assigned or otherwise used by broadcasters or by any other licensed services. With the assignment of channels to broadcasters varying by location, not all the allocated channels are assigned to broadcasters in any given market, giving rise to “white spaces” in which a channel that is not assigned for broadcast may be available for other purposes. TVWSs are a great example of spectrum sharing; regulators allow TVWS devices to transmit on these unassigned channels without the need of a license (unlicensed spectrum). TVWS can be used to provide broadband connectivity and telemetry for Internet of things (IoT) solutions. TVWS can be deployed in conjunction with 5GHz and other mid-band links to increase the coverage area and to cover the missing spots that cannot be addressed by 5GHz and other mid-band links due to non-line of sight (NLOS) and distance limitations.

Vijaya Nirmala Mitnala Single Presenter

There are an increasing diversity of devices and networks for making and receiving video and/or audio calls. The prevalence of smartphones, multiple smart speakers and various computing platforms in many homes has led to much greater flexibility in how and where users communicate. This trend is only set to continue as AR/VR becomes more commonplace and services such as healthcare, for which flexible, high-quality communications is a critical component, move online. Despite these trends, comparatively little has been done to offer a converged communications experience once a speech call session is in progress. For example, the simple act of switching an ongoing call from a smartphone to a smart speaker is often a highly manual process. Pervasive communication systems aim to address this by providing a seamless, flexible communications experience across multiple devices and, where required, multiple networks. A key part of achieving this experience is through seamless session handover. This paper considers how session initiation protocol systems can operate in a pervasive communication scenario, in particular when there is mobility causing vertical handover between delivery networks. The paper uses a machine learning-based approach to predict users\\\' transitions and thus overcome interruptions in speech due to signalling handover. As pervasive communication systems are not currently available for measuring the performance of the solution, the paper uses a commonly available dataset for vehicular mobility to assess likely handover performance. The results show that prediction can reduce the more concerning interruptions (>2s) due to vertical handover events by more than 99.9%.

William King Single Presenter

Over the past decade, there was a gradual trend toward flexible work-from-home schedules. Innovative companies took the risk and became early adopters of the innovative architecture that makes real-time communications possible globally. But the pandemic compressed the normal cycle—which usually transitions from innovative early adopters to becoming a global standard over many years—to weeks. Never before has there been such a fast demand for architectural change. Unfortunately, global network quality can’t match the demand for Global RTC. Common complaints of call abandonment, frozen video, choppy calls are the norm. Leveraging our experience serving some of the largest game companies in the world, Subspace has built the expertise needed to provide a network truly optimized for real-time. This presentation explores the techniques Subspace implements to mitigate network congestion, including specific techniques used for network optimization, and the results realized by implementing those techniques.

Anne Lee Single Presenter

DARPA defines three waves of AI. The first wave is based on handcrafted knowledge. The implementation is symbolic AI which lends itself to basic reasoning. This includes expert systems and some may argue decision trees. The second wave of AI is based on statistical learning and its strengths lie in the ability to learn without being explicitly programmed. The challenge with the second wave is transparency and brittleness. Second wave AI agents are mostly black boxes. Users of a second wave AI agent may question how the AI drew its conclusions. After all, correlation does not imply causation. While the second wave of AI has shown itself to be very powerful, there is much that can be improved.The third wave of AI marries the reasoning abilities and ambitions of the first wave with the learning capabilities of the second wave to produce a solution that is able to generalize beyond the training dataset and can explain itself thus making it more trustworthy and robust.This talk will explore the third wave of AI through a Causal Lens and how Telecom can benefit.

Altanai Bisht Single Presenter

With the spike in WFM and remote learning, there has been a many fold increase in usage of VoIP especially in conferencing ( 20+ users). This talk focuses on the impact of CPU performance on the carbon and greenhouse gas emission from the grid based electricity generation perspective. It will also touch upon BTM(Behind the meter storage) for ToU(Time of Use) billing, peak shaving algorithms and RTC optimizations that yield better energy savings. It is aligned to my ongoing research on whether renewables and energy storage integration green the electric grid. Experimental findings using Call traffic, CPU utilization on various RTP topologies and SIP/WebRTC call flow optimizations will be shared and demonstrated.

Bob Friday Panelist
Dean Bubley Moderator
Edgar Figueroa Panelist
Sander Rotmensen Panelist

Nature abhors monocultures. They're risky and vulnerable. The impact of any catastrophic risk is magnified hugely. They are susceptible to parasites, infections or natural disasters. They can also monopolise resources, driving out other creatures and limiting the potential for further evolution. Biodiversity helps resilience and also leads to amazing “generativity”, as new species – and new interactions between species – enriches the wider environment, as well as long-term evolution.The same applies to technology monocultures – and in this case, wireless connectivity. Listening to some commentators – especially if they’re aiming at radio spectrum rights or government subsidies – you could imagine that 3GPP’s 5G technology could run the whole world, from the phones, to home appliances, to factory automation systems. Yet not only is that improbable, the vision and hype itself risks crowding out competing and interacting technologies, each with strengths and weaknesses. 5G, 4G, Wi-Fi, satellite connectivity, Bluetooth, LoRa, Thread, free-space optics and a plethora of others are all relevant both to users and to help maintain a strong and rich ecology of "network diversity". Sometimes these alternatives can be combined, other times it makes sense to have independent systems. Users, policymakers, vendors, investors and academics should be wary of the elegant – but flawed – concept of the “5G monoculture”.

Alex Lartey Single Presenter

Robocalls are a primary source of consumer complaints to the FCC, making up 60% of all consumer complaints filings. By some measures, up to 50% of all inbound cell phone calls are robocalls or spam. In February 2020 alone, Americans received 5.2 billion robocalls, which works out to nearly 2,000 robocalls per second. Due to this ongoing problem the Federal Communications Commission refers to as "the robocall scourge," 61% of people in the US regularly ignore calls.Governments and tech companies globally are trying to address this problem, but there’s no viable remedy yet. Spam blocking apps that access “black lists” and block calls from unwanted numbers, call screening apps that find out the name of the caller and notify users with corresponding audio prompts are some remedies with mixed results. More recently, major US network operators have implemented call signing through the adoption of the STIR/SHAKEN framework, which is likely to offer a measure of relief, but unlikely to provide consumers with a magic bullet to combat this problem. Thus far, Personal AI Assistant apps appear promising. A few years ago Google added a call screening feature to their Pixel phones, and later went on to expand the call screening functionality to Google Assistant and Duplex. However, this functionality is only available in a few countries. and, those apps have their own limitations: besides spammers, there are missed calls, or times when users simply can’t pick up.As a leading vendor of Conversational AI tools and technologies, Just AI would like to share some first-hand insights on: The benefits of adding NLU technology to build a more effective overall solution with complex functionality: combatting spammers, handling calls, taking messages.

Mathew Lightman Single Presenter

Great datasets are not just born but made and curated. Learn how we are doing more than just taking public audio and text datasets to train our speech recognition models. See what advantages we find with our more detailed approach to creating datasets and how we are building a dataset library for the future of AI voice technology.

Marius Waldum Single Presenter

Why has there been so much more innovation in the web- and mobile than other industries? One reason is that the internet and the two major mobile platforms have been great at enabling third party developers to build a wide range of products, as well as offered a great distribution channel towards end-users. We believe this multisided marketplace approach is the next step for the telecom industry as well.WGTWO has created a programmable mobile core network, built as a platform, API’ed and delivered as-a-service, along with marketplaces where operators, developers and end-users interact. This enables 3rd party developers to create great products that people love - helping operators to start using and redistributing these products - and allowing subscribers to find, add, and manage extra products on top of their price plans.Making this a reality is not straight forward. Let us share our journey, where we started - where we are right now - and the direction we are heading.

Sam Davidson Single Presenter

We explore the application of state-of-the-art NER algorithms to ASR-generated call center transcripts. Previous work in this domain focused on the use of a BiLSTM-CRF model which relied on Flair embeddings; however, such a model is unwieldy in terms of latency and memory consumption. In a production environment, end users require low-latency models which can be readily integrated into existing pipelines. To that end, we present two different models which can be utilized based on the latency and accuracy requirements of the user. First, we propose a set of models which utilize state-of-the-art Transformer language models (RoBERTa) to develop a high-accuracy NER system trained on custom annotated set of call center transcripts. We then use our best-performing Transformer-based model to label a large number of transcripts, which we use to pretrain a BiLSTM-CRF model and further fine-tune on our annotated dataset. We show that this model, while not as accurate as its Transformer-based counterpart, is highly effective in identifying items which require redaction for privacy law compliance. Further, we propose a new general annotation scheme for NER in the call-center environment.

Chris Allen Single Presenter

How can you architect a real-time interactive broadcasting solution? Use cases such as e-sports, gambling, auctions, concerts, and more all call for technical architectures which combine the best of P2P WebRTC and broadcasting technologies in a single solution. Red5 Pro CEO Chris Allen will discuss the evolution from P2P networks to SFUs to what he refers to as the XDN - Experience Delivery Network. Whether your application deals with hybrid events, online education, or security cameras, there are many possible applications for the multidirectional content delivery networks that Chris will speak about.

Toshish Jawale Single Presenter

Human conversations are everywhere, and conversation intelligence is expanding into identifying more and more aspects of the conversations while it\\\'s in progress. Real-time intelligence is unlocking use cases that were never possible before. A Zero-Shot Learning enables the entire ML cycle of learning -> modeling -> Inferring in milliseconds, no training data, training cycle required which takes days if not weeks to build a single model. With the power of Zero-Shot Learning scaling up the conversation scenarios is a matter of few hours instead of weeks.The presentation will cover - * deep-diving into how Zero-Shot learning solves various conversation intelligence problems. * How the technology works with so little latency to build production-ready AI models with architecture. * Demonstration of one example to show it in action. * Possibilities and next steps of what other problems can be solved in completely different ways with Zero-Shot learning techniques.

Henning Schulzrinne Single Presenter

Robocalls are not nearly as dangerous as a pandemic, but, like other security-related threats, they share attributes with communicable diseases. Robocalls have made one of our earliest social media, the phone, much less useful, and caused significant financial losses, particularly to vulnerable populations. Beyond individual measures, we now have some \\\"public health\\\" approaches that may make robocalls harder to monetize and easier to contain. I will provide an update on proposed mitigations for robocalls, their promise and limitations. In particular, I will discuss how mandatory call signing, via STIR/SHAKEN, is likely to play a crucial role, but is unlikely sufficient to make the phone safe to answer again. I will also briefly compare the on-going fight against spam email with the effort to reduce robocalls.

Jason Beloncik Moderator
Nilesh Gandhi Single Presenter

Voice biometrics offers improved account access security and user experience. Modern voice biometrics solutions resist replay attacks and have improved in accuracy and performance making them well suited for call center security and fraud management use cases. Rather than rely in static credentials, modern voice biometric solutions can be used as part of passive and continuous authentication sessions. Continuous authentication is merely a continuation of the voice biometrics authentication. As the customer keeps conversing with the CSR, the voice biometrics solution constantly monitors the customer’s voice.Daon’s presentation will examine the state of voice biometric authentication in the market today. What techniques and technologies are being deployed to increase the accuracy rates, and lower the spoofing rates of voice as a biometric. How voice biometrics can also support virtualization of call centers that also ensures the CSR/employee is who they claim to be. The session will also discuss:• the technical use of metadata and signals to augment how voice is most effectively used as a biometric • app-based versus IVR based voice authentication considerations • how VOIP, SAML and Tokenization can be used to provide greater security and lower operating costs of call centers

Charles Eckel Single Presenter

The User Network Interface (UNI) Manager project in OpenDaylight provides data models and APIs that enable creation and management of carrier ethernet services defined by MEF in physical and virtual network elements. In this session, Charles provides an overview of UNI Manager (UniMgr), the collaboration between MEF and IETF on its YANG models, and the use of OpenDaylight as an implementation platform. In addition to a demo of the functionality, Charles highlights the changes in MEF and IETF that enabled this collaboration and the benefits of combining standards and open source to the OpenDaylight community.The audience is network engineers and software developers involved in standards organizations or open source efforts related to network programmability. It is targeted more toward those working in the service provider space, but the session is generally applicable to other areas as well. The audience will gain sufficient background to understand and appreciate the revolution in internet and service provider networking standards that IETF and MEF are driving as well as how these standards organization are reinventing themselves and embracing open source to remain relevant. Perhaps most importantly, they will learn how and why to engage with such standards organizations to increase adoption of their open source project and gain valuable insights and improvements from a new community of developers and users.