Lorenzo Miniero Single Presenter

Cascaded conferencing has been around for quite some time already, especially in the context of audio mixing. The main idea at the time was linking separate conference bridges somehow, in particular at the mixing level, so that they could cooperate and interact to help scale conference rooms. While the world of communications has changed considerably since then, the same fundamental concepts can still help WebRTC applications, even when using SFUs rather than mixers. In fact, cascading SFU instances can help distributing injected streams across multiple WebRTC servers, which can be useful in different ways: it can help scaling the distribution of streams to a wide audience (e.g., large webinars), making it easier for users to connect to the server closest to them even when it\\\'s not the one hosting the room (e.g., geodistribution), or more in general help implementing applications that may benefit from taking advantage of more than one server at the same time, whether those applications are conference related or not.This talk will present the work that was made in that direction for the open source Janus WebRTC Server, starting from how such a functionality was handled up to now (e.g., to facilitate large scenarios like IETF meetings), and how that had to be changed to acconodate support for PeerConnections with multiple audio and video streams.

Dieff Vital Moderator
Giacomo Paolini Panelist
Shubhendu Bhardwaj Panelist
Valentina Palazzi Panelist

Internet of Everything (IoE) has been a topic of great interest for the past decade. This collective network of connected things 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. Recently, the 5G communication standard has been largely integrated with IoE to further connect things in wearables, tactile internet, vehicles (vehicles to everything), massive IoT (connecting everything), massive MIMO, drone-aided wireless communications, etc. 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 5G/IoE 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 Valentina Palazzi, Giacomo Paolini, and Shubhendu Bhardwaj who will give an industrial and academic perspectives on the research that has been undertaken on IoE and its integration with 5G technology.

Fred Posner Single Presenter

The Kamailio project recently celebrated it\\\'s 20th year and continues to be a dominant force in open source RTC. From security to scalability, SIP to WebRTC, Kamailio continues to integrate new features and remain 100% open source.

Jerod Venema Single Presenter

Brief:As the metaverse takes shape, there’s no doubt augmented reality (AR) and virtual reality (VR) will play a crucial role in how users engage with it. To overcome the challenges of AR and VR, such as device limitations and feelings of isolation, WebRTC offers solutions.Overview:The metaverse means many things to many people, but I’d like to focus specifically on two areas: augmented reality and virtual reality. If we are to build a thriving metaverse, we need to tackle the unique challenges of AR and VR and seize their unique opportunities.Augmented RealityIn augmented reality, we face device limitations. These include minimal input controls and low processing power, but network connectivity has become more reasonable thanks to mobile phones. Here’s how WebRTC specifically helps solve these limitations:- For input controls, why not let a remote party help guide you? Or use artificial intelligence to translate speech to text, and take action that way. - For low processing power, a combination of on-device limited capability and power detection with higher quality remote streamed detection can create a powerful combination. This is especially handy when analyzing the AR environment for free-form objects (i.e. not QR codes and such but more dynamic object detection). WebRTC’s ability to run on these lower power devices and also natively in a browser means multi-party experiences are easy to create and reduce the barrier to entry in AR.Virtual RealityFor virtual reality, the worst result is feeling isolated. While VR is an inherently isolating experience, WebRTC can provide a multi-party experience based on standards that allow interoperability between not just other VR devices, but also AR devices, web browsers, and mobile devices. This brings the greatest possible number of people together to create a social VR experience—taking VR from a solo toy to a shared and present experience.

Bernard Aboba Panelist
Jan-Ivar Bruaroey 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 build on last year\\\'s session on WebCodecs by introducing the WebTransport API. Using demos, we will illustrate how the new APIs can be used to develop exciting new applications.Last Year\\\'s WebCodecs Session: https://www.rtc-conference.com/2021/presentation/?hid=880

Charlie Catlett Single Presenter

In the decade since Chicago's Array of Things (AoT) initiative began deploying IoT devices with edge computation capabilities, nearly every aspect of the systems has significantly advanced -- from processor speed to memory to storage to communications. Even more significantly, the software capabilities embodied in new artificial intelligence frameworks and algorithms have catalyzed new applications for measuring urban and environmental dynamics. Catlett will discuss AoT's successor, the NSF-funded SAGE initiative, which has transformed the underlying platform (Waggle) used by AoT into an open, modular, reproducible experimental cyberinfrastructure has enabled a growing number of scientific communities to design and deploy AoT-like systems.

Vikram Iyer Single Presenter

Dense IoT sensor networks are promising to monitor pollution at fine spatial and temporal resolutions, yet few cities have actually implemented such networks due to high costs and limited accuracy. We address these barriers by developing an end-to-end framework for urban air quality sensing with minimal infrastructure requirements and a custom wireless low-cost sensor. This talk presents Eclipse, a platform for low-cost urban environmental sensing using solar-powered and cellular-connected devices, and results from our deployment in the city of Chicago.

David Zhao Single Presenter

WebRTC gets compared to Zoom quite often, especially around the perceived quality of experience. While this is not an apples to apples comparison, it has not been easy to match the end-to-end efficiency that Zoom brings, especially in multi-user conferences.As a protocol, WebRTC is capable of many such advanced features. However, it's not obvious to most application developers how to take advantage of these capabilities. With a tight feedback loop between the client and the SFU, it is possible to achieve quality of experience that match Zoom's. In this talk, we’ll go into detail about how LiveKit uses adaptive streaming and dynamic broadcasting to deliver end-to-end efficiency.

Fermin Fernandez Single Presenter

Learn how this innovation space in Chicago host a focused IoT Community where you can get inspired and learn from like-minded organizations and connections.

Damien Stolarz Single Presenter

WebRTC has emerged as the primary protocol for the most demanding, ultra real-time video streaming scenarios, such as telepresence, conferencing, surveillance, and drone control. It has also found massive adoption in the media & entertainment industry, both as a production collaboration tool and for streaming live events. In this session, we will go over how WebRTC is being used to enable highly distributed industry workflows, specifically live concert broadcasts, monitoring on-set cameras, virtual set productions with Unreal Engine, and remote VFX and editing. We will also share lessons learned adapting WebRTC to high-frame-rate (60fps), high-resolution (4K) and high-dynamic range (HDR) video streaming applications.

Cody GrandPre Single Presenter

With more than 85% of all 911 emergency traffic originating from cellular devices, the need for location based additional data could not be greater. As more and more commercial enterprise networks establish connectivity into the world of IOT, the ability to tap in to extraneous data from these networks and their devices has never been greater. In many cases the initial alert of an emergency event may originate from a personal device as an incident is witnessed by an individual. The correlation of these session initiations can easily be coupled with the vast amount of IOT data available at the same time. Public Safety agencies could not possibly manage the monitoring of all of the available commercial networks for out of specifications and parameters, however understanding where these networks are, as well as the information contained within them, make a valuable and tempting resource for public safety to query allowing them to put additional eyes on the scene of a potential emergency event. This presentation and demonstration will focus on real-world capabilities to collect, ingest, and report into the ESInet real world IOT data, ultimately available directly to the first responder over the PSBN.

Muhammad Hamza Nasir Single Presenter

WebRTC can often be overwhelming for developers not accustomed to working video/audio. So using CPaaS platforms is their go to because of the ease of quickly getting video integrated with their application. However because of all of the layers of abstraction it might seem that there's no room for optimizations to be done but you'd be surprised about all the room for improvement. That is what we'll explore in this presentation with one of the most popular CPaaS platforms, Vonage Video API's.

Budge Currier Single Presenter

As many states embarked on next-generation 911 deployment, California was clearly one of the first, and indeed most significant undertakings. As progress an expansion continues, the session will take a look back on the success achieved, and set the stage to continue to look forward, expanding services, and adding additional connectivity that brings value to the constituent as well as the first responder.

Chris Allen Single Presenter

Interactive watch parties provide an ideal platform for OTT broadcasters allowing for viral sharing and quickly building out a loyal fan base. Learn how the Red5 team in conjunction with Oracle Cloud , Minami Design, and AMD built the open source solution using Red5 Pro’s cloud based Ultra Low Latency streaming platform as the foundation for engagement. This experience drives new monetization streams, including targeted advertising and sponsorship revenue. Watch parties can ultimately evolve into sports betting opportunities, driving new revenue streams as well. The presentation will cover the WBC/Vive TV Network Watch Party Experience (WPE) which is a solution built with partners Red5, Skreens, Oracle and AMD and enables celebrity appearances like Floyd Mayweather or Mike Tyson to make guest appearances in the watch parties.Red5 CEO and Co-founder Chris Allen will walk participants through the architecture, share a demo of the experience, and discuss what the future holds for next generation sports spectatorship.

Steven Raucher Single Presenter

Placing applications in the cloud was initially viewed to be as simple as hosting an application in a distant data center. Reality has taught us that the cloud is much more than a public data center running on a server farm in the desert. Applications need to be cloud native to maximize all of the efficiencies a cloud architecture brings to the table. While many have dreamed of it, one state actually implemented it. Join RapidDeploy as they review the infrastructure bill for the state of California\\\'s new statewide deployment of end-to-end NG911.

Anne Lee Single Presenter

Today, artificial intelligence has become a mainstream topic in the telecom industry. Standards bodies, such as 3GPP, are designing in support for AI as part of next generation system architectures. At the same time, Responsible AI has become a major topic in the artificial intelligence community as well as governments around the world.Responsible AI addresses the topics of transparency, explainability, safety, security, privacy, reliability, fairness, sustainability, and accountability. Addressing the issues around these topics is critical. For example, in the area of sustainability, recent studies show that the compute resources – hence carbon footprint – required for AI training and inference can be staggering.This talk will describe the components of Responsible AI and how these apply to Telecom. Additionally, this talk will discuss the benefits of some of these components beyond the goals of Responsible AI.

Brooks Shannon Panelist
James Kinney Panelist
Jeanna Green Panelist
Mark Fletcher Moderator
Wolfgang Kampichler Panelist

Industry Collaboration Events (ICE) sponsored by the National Emergency Numbers Association (NENA) provide a testing environment in which vendors and carriers test the interoperability and end to end behaviors that are critical for a next generation 911 deployment. The National Emergency Numbers Association has offered these events to its members since 2009. More information about these events can be found at NG911_ICE.This panel includes representatives of the vendors, service providers and standards bodies that collaborate to make these events an essential venue for ensuring the functionality, security and reliability of these life-saving networks. The test events enable testing of NG-specific functions, interoperability and end to end behaviors. Panelists will describe the testing itself as well as the value that it has added to their development and deployment of the elements that provide the core services to Next Generation Emergency Services IP Networks. Focus will be on new functions to be tested for the first time this year and on what will be upcoming in future ICE events.

Dr. Ali Gaber Mohamed Ali Single Presenter

Cognitive radio can be considered as a viable frequency access framework that overcomes the disadvantages of the licensed-based transmission procedure by allowing the secondary users to access the spectrum to transmit their data. However, due to the evolution of latency-sensitive real-time communication applications such as gaming and extended reality, it becomes more vital that cognitive radio networks should be studied with latency requirements on the data transmission. Recently, Age of Information has introduced itself as a important metric for evaluating the freshness of the transmitted data. In this paper, we investigate the latency and stability analysis of a two-user cognitive radio network that consists of one primary user, one secondary user and their destinations. The latency requirements of the transmitted data packets are taken into consideration by imposing Age of Information constraints on the data transmission of the users. We present two optimization problems, in the first problem, the secondary user stable throughput is maximized under an Age of Information constraint imposed on the data transmission of the secondary user. While, in the second problem, we maximize the stable throughput of the secondary user with respect to Age of Information constraints set on the data transmission of both the primary and secondary users. The resultant problems are found to be non-linear programming optimization problems. An appropriate algorithm is used to solve the problems and provide the numerical solutions. Our results characterize the impact of setting Age of Information constraints on the stability region of the network; we demonstrate that the stability region, in certain cases, is reduced by only 11% with strict Age of Information restrictions if compared to the scenario where no latency requirements is considered. Our results also show the potential accuracy of the algorithm adopted in this paper to solve the formulated optimization problems.

Chuck Spalding Single Presenter

This session will review the various types of cyber attacks affecting ECC’s. Attacks against both 9-1-1 and admin lines will be discussed. Examples of hacked business phone systems and the use VoIP from outside the US will be explained. The primary attack vector for malware to infect CAD and Record systems will also be discussed as well as the impact of NG911 connectivity exasperating the problem if not properly addressed.

Aurélien Agniel Single Presenter

The introduction of the IoT challenges this notion by placing computing resources and observation at the edge of the network. As a result, decentralized approaches for information processing and gathering can be adopted and explored. However, this shift greatly expands the network footprint and shifts traffic away from the center of the network, where observation and cybersecurity monitoring tools are frequently located. Further, IoT devices are often computationally constrained, limiting their readiness to deal with cyber-threats. These security vulnerabilities make the IoT an easy target for hacking groups and lead to the proliferation of zombie networks of compromised devices. Frequently, zombie networks, otherwise known as botnets, are coordinated to attack targets and overload network resources through a Distributed Denial of Service (DDoS) attack. In order to crack down on these botnets, it is essential to develop new methods for quickly and efficiently detecting botnet activity. This study proposes a novel botnet detection technique that first pre-processes network data through computer vision and image processing. The processed dataset is then sent to a neural network for final classification. Two neural networks will be explored, a sequential model and an auto-encoder model. The application of image processing has two advantages over current methods. First, the image processing is simple enough to be completed at the edge of the network by the IoT devices. Second, preprocessing the data allows us to use a shallower network, decreasing detection time further. We will utilize the N-BaIoT dataset and compare our findings to their results.

Eric Hagerson Panelist
Mark Fletcher Moderator
Michael Hooker Panelist

According to several sources, wireless carriers originate over 80% of 911 calls. Therefore, as the nation transitions to NG911, wireless OSPs will be key stakeholders in the process. As a wireless industry leader, T-Mobile has been at the forefront of implementing NG911. This session will explore the challenges and opportunities that wireless OSPs face in the transition to NextGen emergency communications.

Mikhail Gromov Single Presenter

Continued growth and adoption of the Internet of Things (IoT) has greatly increased the number of dispersed resources within both corporate and private networks. IoT devices benefit the user by providing more local access to computation and observation compared to dedicated servers within a centralized data center. However, years of lax or nonexistent cybersecurity standards leave IoT devices as easy prey for hackers looking for easy targets. Further, IoT devices normally operate at the edge of the network, far from sophisticated cyberattack detection and network monitoring tools. When hacked, IoT can be used as a launching point to attack more sensitive targets or can be collected into a larger botnet. These botnets are frequently utilized for targeted Distributed Denial of Service (DDoS) attacks against service providers and servers, decreasing response time or overwhelming the system. In order to protect these vulnerable resources, we propose an edge computing system for detecting active threats against local IoT devices. Our system will utilize deep learning, specifically a Convolutional Neural Network (CNN) for detecting attacks. Incoming network traffic will be converted into an image before beings supplied to the CNN for classification. The network will be trained using the N-BaIoT dataset. Since the system is designed to operate at the edge of the network, it will run on the Jetson Nano for real-time attack detection.

Vijay Gurbani Single Presenter

The systematic monitoring of private communications through the use of information technology pervades the digital age. One result of this is the potential availability of vast amount of data tracking the characteristics of mobile network users. Such data is becoming increasingly accessible for commercial use, while the accessibility of such data raises questions about the degree to which personal information can be protected. Existing regulations may require the removal of personally-identifiable information (PII) from datasets before they can be processed, but research now suggests that powerful machine learning classification methods are capable of targeting individuals for personalized marketing purposes, even in the absence of PII. This study aims to demonstrate how machine learning methods can be deployed to extract demographic characteristics. Specifically, we investigate whether key demographics—gender and age—of mobile users can be accurately identified by third parties using deep learning techniques based solely on observations of the user’s interactions within the network. Using an anonymized dataset from a Latin American country, we show the relative ease by which PII in terms of the age and gender demographics can be inferred; specifically, our neural networks model generates an estimate for gender with an accuracy rate of 67%, outperforming decision tree, random forest, and gradient boosting models by a significant margin. Neural networks achieve an even higher accuracy rate of 78% in predicting the subscriber age. These results suggest the need for a more robust regulatory framework governing the collection of personal data to safeguard users from predatory practices motivated by fraudulent intentions, prejudices, or consumer manipulation. We discuss in particular how advances in machine learning have chiseled away a number of General Data Protection Regulation (GDPR) articles designed to protect consumers from the imminent threat of privacy violations.

Dean Bubley Single Presenter

Ecologically, we understand the benefits of biodiversity. Future network infrastructure should also combine multiple different technological options: fibre, 5G, WiFi, satellite & other options. Both technologists and policymakers should look at avoid single points of failure, or monocultures, and instead foster a wide range of options. These may be both tightly coupled (converged) or separate (diverged) - there are arguments for both. This session looks at ways to exploit and improve network diversity.- Innovation, competition and avoidance of lock-in - Mitigation of supply-chain constraints - Resilience, backup & cybersecurity - Least-cost & lowest-energy routing - Lessons from Ukraine - Shifting roles of service providers, device suppliers and the end user

Thomas McCarthy-Howe Single Presenter

vCons are a proposed data standard that defines an actual human conversation. More than a WAV file, vCons can contain the identity of the participants, the entire media, supporting attachments and analysis, in a tamper proof package. Whether as a critical enabler to control personal information in recorded conversations, or as a standard way to send conversational data into AI and machine learning systems, vCons promote conversations to first class data objects. This session will cover the four main parts of vCons, their relationship to other communications systems, and suggest practical ways of using vCons to improve general business processes.

Utpal Mangla Single Presenter

5G edge computing need to operate across hybrid clouds to be effective but challenges remain. The challenges include the deployment of 5G networks in a hybrid cloud environment and operating a resilient and optimized 5G network across a hybrid cloud environment spanning the public clouds to core networks to MEC compute nodes to far edge devices. This session will discuss the benefits of deploying 5G network with edge computing nodes across a hybrid cloud with relevant industry use cases leveraging AI & Automation. We will also discuss how to effectively deploy a 5G network across a hybrid cloud environment and how closed loop automation built on AI systems integrated with the underlying programmable networking management and orchestration functions can automate in real time Day2 operations of a 5G cloud native environment. The session will also give an overview of the KEY Industry EDGE Cloud Use cases. We will conclude with a demo so that the challenges and solutions to address 5G computing in a hybrid cloud environment are clear to the audience and a discussion on how far along communications service providers are in their journey to automate 5G network operations in a hybrid cloud environment.

Damian De Rozairo Single Presenter

We will demonstrate how the evolution of voice technologies has simplified the user authentication process across all interactions between an individual (employee/customer) and an organization.

Bernard Malone III Single Presenter

Outages from various telecommunications network-impacting events over the last couple of decades provide observations which can help categorize vulnerabilities and provide concepts for improving resiliency and availability of networks in an automated or programmable manner. Network-impacting events can be caused by a variety of events including natural causes, man-made or terrorist induced and geopolitical based. Such events cause service disruptions and outages, can be short or long term, and last until restoration or re-routing takes place. When external events cause network disruptions, users lose valuable communications services. Analyzing the telecommunications model that includes client devices, access, back-haul, transport, core and applications services in several past disasters can highlight vulnerable areas. Studying not only individual disasters, but the larger trend, can help to identify common issues. Designs can then be considered to identify and implement parallel network resources in each of these areas. Identifying these additional layers of network resources can then lead to developing Real-Time Programmable ways to implement them.

Daniel Pluth Single Presenter

While redaction of sensitive information can be a relatively simple natural language understanding (NLU) problem when applied to written word, the application of these same methods to speech is more complicated. Typical approaches rely on automatic speech recognition (ASR) to reduce the problem to text, which may fail to completely redact information due to ASR errors. We propose an end-to-end model, applied directly on speech, in order to create an appropriate mask on sensitive information.

Dinesh Verma Single Presenter

Quantum Communications Networks have their unique communications characteristics and have the potential to enable significant support for both traditional computer applications as well as quantum computer applications. Although still in experimental stage, they are likely to become a reality in the near future. Developing a control software for quantum communications networks, which allows the quantum communications networks to complement the capabilities of a traditional communication network can yield a significant number of new applications and solutions. In these configurations, a quantum network complements a communications network, enabling two channels of communication between any two nodes. By making the control software of this network programmable, the operation of both channels can be improved. IBM Research and Brookhaven National Laboratories are exploring the development of a common control software for quantum communications networks. In this talk, we will discuss that activity, and explore how adding programmability to the control software can be valuable in the context of such multi-channel networks.

Diana Deibel Panelist
Nancy Munro Moderator
Shyamala Prayaga Panelist
Susan Westwater Panelist

The majority of STEM jobs are still filled primarily by men. The exception seems to be voice technology. Join our panel of women voice technologists and designers to hear their stories of how they got to where they are today, where voice technology is going, and what types of skills and educational paths will help open the door for others seeking to hop on this growing field of voice technology.

Varun Singh Single Presenter

After ten years of slow and steady increase, interest in WebRTC has been massively accelerated. The onset of the pandemic rocketed the need for real-time collaborative experiences at both small and grand scale, while at the same time increasing the world’s comfort level with the technology. We have seen a huge rise in productivity, collaboration, and social environments mapped onto social spaces.With regard to the future of WebRTC, discussion included:What are people willing to pay for (monetizable) at scale at different points of a technology evolution? Which are the WebRTC use cases that don’t need more innovation and which will become commoditized? Where is there still room for creativity?Do we want to duplicate the experience when building for remote events, or enhance it? What experience will we be optimizing for? The most valuable version of the experience may be the digital version – if we get it right.Creator-driven content will continue to rise – live commerce, concerts, social. So will the need for highly produced video experiences made possible by great creator tools. At every layer, we’re actually still very early in the evolution of where the use cases for WebRTC can and will go. Every space is going to grow and WebRTC is a core building block for that future.According to us, what really matters in WebRTC development is: * Developer time to value * Call quality and reliability * APIs for the largest possible set of use cases

Karl Stolley Single Presenter

This talk will offer attendees an example-driven look at the often under-explored `RTCDataChannel` API. Given all of the challenges and interest around audio and video, data channels find themselves as the short leg of WebRTC’s three-legged streaming stool.The talk will open with a closer look at the range of options available when setting up a new data-channel instance with the `RTCPeerConnection`’s `createDataChannel()` method. The available options will be examined against both browser defaults and browser support—with special attention to the lingering browser-implementation issues surrounding the `Blob` value on `binaryType`.From there, the talk will shift to applied techniques for coordinating multiple data channels within a single WebRTC app. The example provided attendees will be data-channel only, and it will demonstrate the use of multiple data channels working independently of and in concert with one another. The example, which the talk will walk through, will include negotiated and non-negotiated (out-of-band) data channels, data channels that are short- or long-lived, and various strategies for routing incoming data-channel events and data to a specific callback using both data-channel `label` values, as well as the user-definable `protocol` property.The talk will conclude with a look at data-channel features being discussed for the next iteration of the WebRTC specification, especially a proposed `getDataChannels()` method. That method is especially needed, because currently there is no API-native method for retrieving all of the data channels attached to a particular `RTCPeerConnection` object. That predictably places a lot of cognitive load and overhead on developers, who must themselves keep track not only of which channels are opened, but what each channel’s current `readyState` is.

Ishrat Ahmed Single Presenter

In this project, we focus on developing a model, based on pre-trained open-source transformer-based language models (LM), to process a customer’s phone call or voicemail transcript and generate a summary. While LMs such as GPT-3 can generate a reasonably accurate summary of the transcript of a phone call or voicemail,  integrating GPT-3 into a commercial service raises cost and privacy concerns. So, this project aims to experiment with a few open source LMs for text summarization, and finetune a subset of them with our own dataset, then evaluate the performance using commonly adopted metrics such as ROGUE score, so that the best performing model for our dataset can be obtained for potential use in analyzing call transcripts.

Chad Hart Single Presenter

WebRTC was obviously popular with developers during the peaks of the pandemic, but how is it doing now? Did all those new projects die, putting the community back at pre-pandemic “normal” levels or is WebRTC still going strong? Are there many new WebRTC-related repos? Is WebRTC still attracting new users and what are they doing? How are newer API’s like Insertable Streams and WebCodecs doing?In this talk, Chad Hart of webrtcHacks and Kranky Geek will answer these questions with a fresh analysis based on a review of over a million GitHub events since 2019.