Webinars, Events and Conferences

Register and watch Omniscien and partner webinars. If you miss a webinar or cannot watch it live at the scheduled time, come back after the webinar and watch the replay any time.

Stay tuned. We will be announcing new webinars shortly. Subscribe to our mailing list to be notified.

Please check out our Past Webinars in the meantime or subscribe to our mailing list if you would like to be notified by email of our new webinar dates.

We also have a large number of webinars available at our annual Artificial Intelligence, Machine Translation and Language Processing Symposium website.

AI and Language Processing Innovations

What Is It Good For? Real-World Use Cases

Monday 22 April 2024

As many are intrigued by AI but unclear about its real-world utility, this session will showcase groundbreaking use cases, revealing AI’s tangible benefits across diverse sectors through the lens of language processing. This webinar delves into the advancements and applications of AI and language processing technologies, highlighting their potential to revolutionize various sectors and the impacts they are already having. Designed to demystify AI’s practical applications, it offers a deep dive into how these technologies are shaping our future.

Join Omniscien’s Chief Scientist, Professor Philipp Koehn, and CTO, Dion Wiggins, for an in-depth look at practical AI applications across various industries, with a focus on language processing and its real-world utility. Discover how these technologies are currently transforming businesses and services.

Key Topics:

• Advancements in AI and Language Processing: An overview of the latest AI tools and language processing technologies and trends, with insights into AI adoption across various sectors and its impact on market dynamics. Explore AI’s role in transforming workplace productivity and enhancing skills, highlighting its integration into the future of work.
• Innovative Use Cases: A walkthrough of a series of innovative use cases that demonstrate AI solutions to real-world problems.
• Language Studio Demo: Experience Language Studio’s capabilities in action, including translation, speech-to-text, and text-to-speech applications. Additionally, see live demonstrations of generative AI models and NLP solutions that enhance decision-making and operational efficiency.
• Q&A Session: Interactive session with the experts to address your specific questions.

Who should attend? This webinar is intended for professionals looking to integrate AI into their operations, policymakers, and anyone interested in the practical impacts of AI and language processing technologies.

Participate in this session to gain clear, actionable insights into leveraging AI and language processing innovations in real-world scenarios.

Speakers:

Advances in Artificial Intelligence and Language Technology: 2024 and Beyond

Demystifying Large Language Models (LLMs) and more.

Monday 4 December 2023

In the past 18 months, we have witnessed an unprecedentedly rapid transition and adoption of technology. Artificial Intelligence (AI), shrouded in both hype and mystique, has made a profound impact across various industries.

This webinar explores AI’s influence on language processing, workflows, and data processing within businesses. Our goal is to demystify AI in the realm of language processing, providing clarity on its real-world implications, risks, and benefits.
 We will assess AI’s current state and forecast its trajectory into 2024 and beyond. The webinar will cover significant advancements in AI as it relates to language, including Machine Learning (ML), Automated Speech Recognition (ASR), Machine Translation (MT), Media Processing, and Natural Language Processing (NLP).

We are excited to welcome Omniscien’s Chief Scientist, Professor Philipp Koehn, Chief Technology Officer, Dion Wiggins, and guest speaker Dr. Joseph Sweeney. Dr. Sweeney is a Research Director and advisor on the Future of Work at Intelligent Business Research Services (IBRS), a leading ICT and digital advisory firm.

New Language Studio Generative AI Module

Additionally, we’ll introduce the new Generative AI Module in Omniscien’s Language Studio. Experience a live demonstration of our cutting-edge Generative AI platform, reminiscent of ChatGPT but enhanced with your data for superior accuracy and control. This latest addition to the Language Studio suite allows enterprises to utilize AI while ensuring privacy and security. Discover how this technology can be integrated securely within your office network. With Language Studio, your data remains solely under your control, securely managed behind your firewall or within your self-hosted private cloud.

Join Omniscien’s Philip Koehn and Dion Wiggins, along with guest speaker Dr. Joseph Sweeney, for a comprehensive review of the rapidly evolving AI landscape from the context of enterprise enablement via AI, best practices, risk management, and future roadmaps.

Speakers:

Guest Speaker Bio

Dr. Joseph Sweeney, is a seasoned advisor and consultant at Intelligent Business Research Services (IBRS), where he specializes in business and technology strategy, emphasizing AI, data privacy, and compliance. He helps organizations navigate the complex landscape of emerging technologies, with deep expertise in education, the public sector, and Microsoft technologies. Dr. Sweeney’s background in AI, cloud computing, and technology trends enables him to offer valuable insights for developing strategic solutions. As a sought-after advisor, he guides organizations through digital transformation, market research, and the integration of cutting-edge technologies.

The Future of Language Related AI for Enterprises:

Local Agents and Fine-Tuned Large Language Models (LLMs)

Thursday 15 June 2023

Artificial Intelligence (AI) has revolutionised how we process and understand human language. At the heart of this transformation are two modern features. First is Large Language Models (LLMs) – powerful AI systems that comprehend and generate human-like text, leveraging deep learning techniques and extensive datasets. The second is Agents that perceive and interact with their environments through natural language processing (NLP) to ask questions, make decisions, produce data, and take actions.

However, the one-size-fits-all approach of LLM training, seen in OpenAI’s ChatGPT and Google’s Bard, often constrains their proficiency in task-specific efficiency and organisation specific knowledge, while also being fraught with data privacy and compliance risks.

The solution? AI models refined by cost-effective fine-tuning tools that cater to unique application requirements, empowering businesses to swiftly modify AI models with their proprietary data and process requests securely within the privacy of their own corporate networks.

Secondary to the solution is utilising natural language APIs atop tailored AI models, where secure self-hosted agents can interact with a curated, expandable toolset that excels in multimodal tasks, generating images, summarising text, transcribing and analysing audio and video, reading aloud, answering questions based on documents, generating company specific text, articles, data, and many more NLP features.

In the face of the current deafening level of AI noise from media, social media and vendors, charting the path for enterprise AI may seem daunting. This webinar aims to demystify the current state of relevant technologies. Omniscien’s AI experts will articulate their vision of a bright enterprise AI future powered by affordable fine-tuning tools for AI models and natural Language APIs, further amplified by the use of secure, self-hosted agents. The session will also showcase some existing and future features of Omniscien’s Language Studio platform as practical illustrations of what’s already achievable.

Join Omniscien’s Philip Koehn and Dion Wiggins for a comprehensive review of the rapidly evolving AI landscape from the context of enterprise enablement via AI, best practices, risk management, and future roadmaps.

Speakers:

Rise of the Machines: Balancing Language-Related AI Opportunities and Risks

Securing the Future of Generative AI, Machine Translation, Speech Recognition, NLP and AI Augmented Processes

Tuesday 25 April 2023

ChatGPT 3.5’s launch in November 2022 transformed the AI landscape by making sophisticated, easy-to-use, and productivity-enhancing AI capabilities accessible to everyone online with little-to-no cost or experience. While related to specialized AI technologies, such as NMT for machine translation and ASR for voice recognition, its accessibility and adaptability set it apart.

The ensuing media spotlight on AI highlighted opportunities and risks, prompting enterprises to seize new opportunities, while governments quickly acknowledged the need for regulatory measures and controls. Italy banned ChatGPT due to perceived risks, and over a thousand stakeholders called for a research halt, citing ethics. As the hype subsides, realism and responsibility are reemerging for enterprises, as exemplified by Samsung after it lost control and ownership of sensitive data,  intellectual property and trade secrets.

In Part 3 of our 2023 AI Webinar Series, we delve into the secure and private use of language related AI technologies and discuss the future of AI for enterprises, how it augments human processes, and where it can be best utilized in a secure and private manner.

We will explore the following topics:

1. Generative Text, Machine Translation, Voice Recognition and Optical Character Recognition in the new AI era.
2. The perceived threats and benefits that AI augmented process bring to enterprises
3. Real-world examples of prominent risks in the latest AI wave
4. How GDPR and other compliance regulations fit with language related AI technologies
5. The need for and expected regulatory changes and their impact on enterprises
6. Strategies for mitigating and managing AI-related risks
7. How technology is already adapting to address these risks and where it is heading in future.

Examples of technologies:

1. Domain adaptation of foundation models
2. Enterprise specific models for greater accuracy and relevance
3. On-premise and on-device AI models
4. Real world examples using Omniscien’s Language Studio AI platform

Join Omniscien’s Philip Koehn and Dion Wiggins, along with guest speaker Dr. Joseph Sweeney, for a comprehensive review of the rapidly evolving AI landscape from the context of enterprise enablement via AI, best practices, risk management, and future roadmaps.

Speakers:

Guest Speaker Bio

Dr. Joseph Sweeney, is a seasoned advisor and consultant at Intelligent Business Research Services (IBRS), where he specializes in business and technology strategy, emphasizing AI, data privacy, and compliance. He helps organizations navigate the complex landscape of emerging technologies, with deep expertise in education, the public sector, and Microsoft technologies. Dr. Sweeney’s background in AI, cloud computing, and technology trends enables him to offer valuable insights for developing strategic solutions. As a sought-after advisor, he guides organizations through digital transformation, market research, and the integration of cutting-edge technologies.

Redefining Domain Adaptation for Machine Translation and Voice Recognition – An Essential Primer

Thursday 16 February 2023

Large Language Models (such as ChatGPT, Machine Translation (MT) and Automated Speech Recognition (ASR) are three of the many different Artificial Intelligence (AI) technologies today that use Transformer models, so their domain adaptation (aka customization) processes have a lot in common. Domain Adaptation (also known as customization) is the adaptation of AI models used in MT and ASR to a specific domain or purpose such as automotive legal documentation, medical discussions, or investment portfolio sales.

This webinar will demonstrate how Omniscien’s team gathers and synthesizes billions of sentences of training data that is used to teach and adapt AI systems. We will explore best practices and explain the limitations of traditional approaches. We will show how to use AI to create many millions of sentences of training data that can be used as part of domain adaptation. Essentially, this means using one form of AI to teach another AI form new knowledge.

Modern AI systems are built on foundation models containing billions of parameters and need modern approaches to creating training data. New approaches are needed. The legacy historical approaches of simply uploading a few thousand sentences of data are insufficient. Some simple examples are listed below:10K-500K sentences that an organization can typically provide are insufficient to have any notable impact against billions of parameters used in modern foundation models.

• Translation Management Systems (TMS) such as memoQ, Trados Studio, XTM, etc., do not store translation memories in optimal formats for training modern AI algorithms. Translation memories are not in sequence of the original document and are de-duplicated making them unsuitable for training document level translation systems where entire bilingual documents in sequence are needed to obtain the correct context.
• Language Service Providers (LSPs) and translation departments may keep translation memories classified by organisation or customer, but seldom by writing style or purpose. Content from sales, marketing, engineering, legal and other areas are all mixed together.
• Glossaries alone are absolute and result in over-fitting in the wrong context. Many MT systems using glossaries simply search and replace rather than use the glossary in context and the use of the glossary is not 100% reliable.

We will also explore how the advantages that LSPs and large organizations have had in the past by having “larger amounts of training data” on file are very quickly becoming irrelevant in the new redefined approach to domain adaptation where tens or evens hundreds of millions of sentences are needed to have any meaningful effect. Fortunately, there are already solutions that address this issue by enabling any organization to create this data quickly and cost effectively.

Consumers enjoy using machine translation, voice recognition, optical character recognition, and other NLP processes using the web, or using their phones, their Alexa or Google Home devices, etc. However, each time data is sent to these cloud services, the providers are using your data for a wide variety of purposes. Thus, what may be great for consumers, is often not practical for organizations due to data privacy and legal compliance issues and regulations (i.e., GDPR, CCPA, Australian Data Privacy Act, HIPPA, etc.). We will also explore how organizations can take advantage of the latest state-of-the-art technologies without losing control of their sensitive data and running afoul of regulations.

The Omniscien team has been hard at work developing our latest versions of market-leading machine translation, media and subtitle processing, and workflow automation products. Currently being finalized and in beta test ahead of their upcoming release, we are excited to announce a series of new products and product updates.

Join the Omniscien team for 3 presentations, including 2 new product preview webinars, and a comprehensive deep dive into the latest advances in data gathering, data synthesis, data cleaning, and best practices for customizing a high-quality translation engine.