Category: Digitization

Preserving Knowledge, Powering the Future: A Conversation with Patrick Fleming, Business Director, Digitization at Ninestars

At Ninestars, Patrick Fleming, former British Library director, award-winning journalist, and digital transformation leader, is helping shape the future of digitization. Drawing from decades at the intersection of culture, media, and technology, he reflects on how digitization is transforming libraries and archives: from fragile collections made searchable and accessible, to AI-powered discovery that redefines how we engage with history. For Fleming, the challenge is not just preserving knowledge but making it intelligent, ethical, and impactful for generations to come.

Over the last few months at Ninestars, Patrick Fleming has been bringing his deep expertise in archives, publishing, and digital transformation to shape our digitization practice. A former British Library director, award-winning journalist, CEO, and consultant, Patrick has spent his career at the intersection of culture, media, and technology. His experience leading transformational programs, from the British Newspaper Archive to large-scale library development initiatives, now informs how Ninestars approaches digitization in the age of AI.

        We sat down with Patrick to talk about how digitization is changing libraries, archives, and the very way we think about access to knowledge.

You’ve worked extensively with libraries and archives over the years. Tell us a little about the transformative shift you’ve witnessed in the move from physical to digital collections.

Technology and innovation has radically transformed library and archive collections. With AI we are entering the next stage of digital innovation.

Digitisation transforms archives and libraries by enhancing access to collections, ensuring preservation of fragile Items, and enabling digital preservation of born-digital materials.

I have been at the heart of this digital transformation at the British Library. The Library’s newspaper collection, possibly the greatest in the world. grows as a paper collection every day due to legal deposit. Enshrined In legislation legal deposit forces the Library to collect the hard copies of every published newspaper In the UK. Only 2% of the collection of over 180 kilometres of content was digitised when the Library introduced its transformative journey to store, preserve and give access to its newspaper collection.

The game changer for the Library was to find a partner who would provide the innovation required to digitise out of copyright newspapers. DC Thomson won a procurement to start the process and the British Newspaper Archive today has 95 million pages and continues to grow.

Has there been a huge change in how people use libraries today compared to when you started?

Libraries and digitisation specialists like Ninestars continue to evolve internationally. Digitisation has transformed researcher behaviour by providing instant access to vast amounts of material remotely, changing research from physical exploration  to online key word searching and data analysis. This shift enables large research questions, facilitates interdisciplinary approaches and poses challenges to the nature of libraries.

Researchers no longer sift through physical documents but rather read materials on screen. Powerful search facilities allow researchers to quickly locate information by searching for keywords across millions of documents. It was painstaking but often failed to capture nuance. A query like “climate change as reported In newspapers before 1988” could return thousands of results, not all of them relevant, now with AI driven digital archives the experiences changes completely. AI models understand context, semantics and intent, instead of matching words, they return answers. They summarise, highlight connections across decades and even suggest related themes. AI In digital libraries adds context, speed, and intelligence. Instead of static repositories, archives become dynamic, exploratory ecosystems as Ninestars are proving with best in-class digitisation transformations. Digitisation has fuelled the growth of digital humanities, creating new Interdisciplinary programs and fostering collaboration between researchers and archives.

How has digitisation changed the role of a librarian? What parts of the job have become easier, and which parts more complex?

Digitisation has transformed the librarians role from gatekeeper of physical collections to facilitator of digital access, shifting focus from preservation to curation and requiring new skills in technology, project management, and Information literacy. Librarians now manage digital resources, develop information literacy programs, teach users advanced search strategies and curate online content. They have become essential educators and technology experts In the digital age.

At Ninestars, we talk about preserving knowledge while making it intelligent and accessible. From your experience, what’s the biggest opportunity and challenges for organizations in embracing deep tech like AI?

AI presents huge opportunities for libraries to automate repetitive tasks, enhance user experience through personalised recommendations and Improved search, support collection development with data analysis, facilitate research with data management and text mining tools and relate dynamic content. Libraries can leverage AI to streamline workflows, optimise resources allocation, and offer accessible services like translation.

What’s one misconception about libraries that you wish more people understood?

Common misconceptions about libraries include:

  • They’re just for books. In reality, libraries also offer technology, music, programs, and services.

  • They’re always silent. Many function as vibrant community hubs with designated quiet zones.

  • Librarians just read all day. In fact, they manage diverse services, collections, and complex databases.

  • Libraries are irrelevant. On the contrary, they provide vital access to expensive databases, reliable internet, and digital literacy support.

Libraries today are dynamic community centers and social anchors, offering far more than print media. They provide learning opportunities, resources, and a safe haven for people of all ages.

What are some of the biggest trends and innovations you see for libraries on the horizon?

AI Is on every Librarian’s lips. Its full scope and International acceptance is still unknown. AI offers Insights for medicine, strategic planning and a host of industry wide uses.

And since we are talking about libraries and archives, is there a book that had a profound impact on you growing up?

To the Lighthouse by Virginia Woolf had enormous Impact on me.

At Ninestars, Patrick’s experience reinforces our belief that preserving knowledge is only half the story. The real transformation lies in making it intelligent, accessible, and impactful for generations to come.

We’d love to hear your thoughts on the future of digitization. Let’s talk at contactus@ninestars.in

The Role of AI in Enhancing Accessibility in Digital Libraries

As digital libraries expand their reach across the world, the mission is no longer just to digitize collections—it’s to make knowledge accessible to everyone, regardless of ability, language, or location. In 2025, artificial intelligence (AI) is emerging as a remarkably powerful enabler of inclusive access in digital libraries. 

Here’s how AI is transforming accessibility from a checkbox into a core design principle. 

Reimagining Accessibility in the Digital Age 

Historically, accessibility in libraries focused on physical access—ramps, large-print books, or audio formats. In the digital era, accessibility extends far beyond and now means: 

  • Navigating complex archives using screen readers 
  • Accessing content in multiple languages and formats 
  • Ensuring metadata and structure support discoverability for all users 
  • Creating a seamless experience for users with visual, cognitive, or mobility impairments 

Traditional digitization approaches often fail to address these needs comprehensively. This is where AI steps in. 

AI-Powered OCR: Making Text Truly Searchable 

Optical Character Recognition (OCR) has long been used to convert scanned documents into searchable text. However, conventional OCR tools struggle with poor print quality, handwritten content, or non-Latin scripts. 

Advanced AI-driven OCR, like the kind found in platforms such as AOTM, offers: 

  • High accuracy across 70+ global languages and scripts 
  • Support for right-to-left (RTL) languages and complex document layouts 
  • Improved recognition of handwritten manuscripts and poor quality texts 

For users relying on screen readers or text-based navigation, this level of precision ensures they can access content that would otherwise remain locked in image scans. 

Multilingual Translation and Transcription 

Language is often an invisible barrier in digital libraries. AI can remove it by: 

Automatically translating content into multiple languages 

Providing real-time transcriptions for audio or video archives 

Generating multilingual metadata to expand discoverability 

By doing so, AI not only breaks down geographic silos—it makes global heritage, research, and literature accessible to linguistically diverse audiences. 

Metadata Enrichment for Better Navigation 

Without strong metadata, even the most valuable documents can remain hidden in digital archives. 

AI tools now enable: 

  • Automatic tagging based on entity recognition and context 
  • Intelligent summarization for quicker content previews 
  • Classification based on genre, era, subject matter, or format 

This metadata isn’t just useful for general users—it’s crucial for those using assistive technologies who rely on structured navigation to understand content and context. 

Text-to-Speech and Audio Integration 

For visually impaired users, AI-powered text-to-speech (TTS) systems provide a lifeline to digital content. Today’s AI models can: 

  • Read scanned books aloud with human-like intonation 
  • Adjust pronunciation based on context (e.g., acronyms vs. abbreviations) 
  • Support regional accents and dialects in multiple languages 

This has opened up new ways for users to engage with archives—whether listening to historical texts or consuming academic papers on the go. 

Adaptive Interfaces and Personalization 

AI allows digital libraries to offer personalized, adaptive user experiences, such as: 

  • Interface adjustments for dyslexic users (e.g., font type, spacing) 
  • Content reflow and contrast enhancements for low-vision users 
  • Predictive search and smart filters based on user behaviour or learning preferences 

In essence, every user can experience a library that feels designed for them—not a one-size-fits-all platform. 

The Road Ahead: AI as an Equity Engine 

As digital libraries continue to grow, AI will play an essential role in: 

  • Democratizing access to knowledge across socio-economic divides 
  • Preserving cultural heritage in native scripts and languages 
  • Supporting inclusive education through accessible digital archives 

To achieve this, institutions must view accessibility not as a compliance metric, but as a foundational pillar of digital transformation. 

Scalable Accessibility Starts with Intelligent Digitization 

At Ninestars, accessibility is built into the foundation of digital transformation. 

 Our AI-driven solutions help digital libraries: 

  • Digitize complex documents with precision OCR
  • Automate multilingual tagging and smart classification
  • Ensure compatibility with assistive technologies
  • Scale accessibility across millions of pages—accurately and efficiently

Building a library for everyone?  Let’s talk about how Ninestars can help you make it possible.  Get in touch at contactus@ninestars.in.

From Braille to Bytes: Digitizing Resources for Visually Impaired Users

From the invention of Braille to the rise of audio recordings, the path to access has always been long for visually impaired readers. Today, digitization for accessibility is reshaping that journey, powering screen readers, text-to-speech, and adaptive formats that make knowledge truly usable. At the heart of this transformation are inclusive digital libraries, where AI ensures content is not just preserved but personalized and accessible to all.

Did you know Louis Braille was just 15 when he invented Braille, the tactile writing system for the visually impaired, in 1824? Despite its transformative impact, fewer than 10% of people who are blind and partially sighted can read Braille. With an estimated 2.2 billion people globally living with vision impairment and 36 millions of them blind, that leaves so many without access to written knowledge.

The Long Path to Access

For centuries, visually impaired readers relied on Braille or audio recordings, both transformative but limited. Braille production was slow and expensive, while audio materials were bulky and scarce. Access to knowledge remained inconsistent for many.

Digitization as a Turning Point

Digitization for accessibility became a game-changer. By scanning books, newspapers, and manuscripts, libraries began converting printed content into digital formats. That made it possible to create screen reader–friendly text, synthesize speech, and adapt documents for visually impaired users. This turning point laid the foundation for inclusive digital libraries. But digitization is only a part of the story. Formats like Accessibility EPUB have been pivotal in ensuring that eBooks themselves are designed to be inclusive. We’ve explored this in detail in our blog Making Digital Journals and Books Accessible with Accessibility EPUB.

AI: Expanding What’s Possible

If digitization was the first step, artificial intelligence has now made those digital archives smarter and more responsive:

  • AI in digital libraries powers OCR that can read faded, historical texts accurately.
  • AI-driven digital archives enrich metadata, tagging names, places, and themes automatically.
  • Natural Language Processing allows summarization and paraphrasing, making content easier to consume.
  • Text-to-speech and language translation open content to global audiences instantly.
  • Generative AI enables conversational search—ask a question, and the archive delivers synthesized answers.

Combined, these tools transform inclusive digital libraries into dynamic, personalized experiences for visually impaired users.

Real-World Change

Some national libraries are leading the way. For example, in India, digitizing Braille books and educational materials has improved accessibility dramatically. Schools and libraries now provide inclusive digital libraries that serve learners more inclusively.

At the same time, global platforms are ensuring that content isn’t just digitized but truly accessible, closing the gap between preservation and usability.

Ninestars’ Role in Shaping the Future

At Ninestars, we understand that accessibility is not an add-on, it must be built in. With decades of experience in digitization, we deliver AI-powered solutions that make collections both preserved and accessible. Our digitization solution ensures clean OCR, structured metadata, and output formats ready for assistive technologies. We help clients build inclusive digital libraries that serve everyone.

What’s Next for Accessibility?

The future of access lies in combining digitization with intelligence. Imagine asking an archive:

“How did people respond to the Great Exhibition in 1851?”

Instead of scanning dozens of pages, a visually impaired student could hear a summary with key documents cited. That’s what inclusive, AI-driven digital libraries enable.

The story of accessibility is ongoing. From Braille to digitization to accessible formats like EPUB, each step brings us closer to a future where everyone has equal access to knowledge. To dive deeper into the role of EPUB in shaping accessible publishing, see our blog Making Digital Journals and Books Accessible with Accessibility EPUB.

Want to know more about digitizing for accessibility? Drop us an email at contactus@ninestars.in

AI and Generative Search: The Next Leap for Digital Libraries

Digitization was only the first leap for libraries. The next frontier is AI and generative search — transforming static digital collections into living, intelligent archives. Instead of sifting through endless results, users can now experience contextual discovery, summarization, translation, and intuitive pathways that make knowledge more accessible than ever.

At Ninestars, we know this future is only possible with strong foundations. Having digitized over 20 national libraries worldwide, we combine scale with precision — from OCR and metadata enrichment to AI-driven workflows that unify global standards. For us, digitization and AI are not just about preservation, but participation — bringing cultural heritage to life for researchers, students, and readers everywhere.

The journey of knowledge preservation has always mirrored the evolution of technology. Stone tablets gave way to manuscripts. Manuscripts were replaced by the printing press. And now, in the digital age, information in libraries and archives are no longer limited by walls or shelves. The knowledge is accessible and searchable from wherever you are.

Digitization was the first leap. Millions of books, manuscripts, newspapers, documents, and photographs were scanned and stored in digital formats, ensuring their survival for future generations. But vast digital repositories alone are not enough if users cannot easily find or interact with them. This is where AI in digital libraries becomes the natural next step.

From Digitization to Intelligence

For decades, researchers relied on keyword-based search to navigate collections. It worked, but often failed to capture nuance. A query like “climate change as reported in newspapers before 1988” could return thousands of results, not all of them relevant.

With AI-driven digital archives, the experience changes completely. AI models understand context, semantics, and intent. Instead of matching words, they return answers. They summarize, highlight connections across decades, and even suggest related themes.

How AI is Transforming Digital Libraries

AI in digital libraries adds context, speed, and intelligence. Instead of static repositories, archives become dynamic, exploratory ecosystems.

  • Smart Search and Discovery: AI understands meaning, not just words. A researcher looking for “climate change coverage in 1970s newspapers” can find relevant articles even if the sources use different phrasing.
  • Contextual Understanding: OCR made text searchable, but AI can analyze themes, relationships, and sentiment over time.
  • Automated Metadata Enrichment: AI extracts names, places, and dates automatically, improving discoverability.
  • Language Accessibility: A 1910 French newspaper can be instantly translated for an English reader.
  • Personalized Research: AI guides users differently—a historian studying migration and a student learning about World War I will each get tailored paths through the same archive.

Generative Search: A Leap Beyond

If AI powers intelligence, generative search brings it to life. Unlike traditional search that lists documents, it creates synthesized answers.

Imagine asking:
“What was public sentiment about railways in 19th century Europe?”

Instead of making the user comb through hundreds of documents, AI-driven digital archives can summarize perspectives across sources and present a coherent narrative. Knowledge becomes conversational, not static.

The Next Step After Digitization

Digitization laid the foundation. Clean scans, OCR, article segmentation, and metadata enrichment make the application of AI feasible. Ninestars has deep expertise in these building blocks, perfected while working with leading institutions like the National Library of Australia and the Royal Danish Library. Large-scale programs, processing over 11 million pages in Australia and 32 million in Denmark, prove that scale and accuracy go hand in hand.

Once digitized, the libraries can prepare the collections for AI in digital libraries. Poor-quality scans or inconsistent metadata can limit the application of AI, which is why digitization and intelligence must go together.

How Ninestars Helps Libraries To Integrate AI Pre or Post Digitization

At Ninestars, we see digitization and AI as inseparable. Our Intelligent Automation Platform (IAP) already uses AI for OCR, metadata tagging, and automated quality checks. We are also building solutions that make archives AI-ready, including:

  • AI-enhanced OCR and content structuring
  • Metadata enrichment powered by machine learning
  • Cloud-native workflows ready for integration with generative search tools
  • Future-ready archives designed to adopt evolving technologies

For libraries and archives worldwide, the opportunity is clear: digitize today, and prepare for an AI-powered tomorrow.

What Generative AI Means for Users

For students, it means a shortcut to discovery—clear, contextual summaries instead of endless lists. For historians, it surfaces forgotten voices in millions of pages. For casual readers, it creates intuitive pathways through culture and history.

This is the true promise of AI in digital libraries: turning preserved knowledge into active discovery.

Challenges Along the Way

AI is not magic. Damaged documents, faded text, or unusual typefaces can complicate results. High-quality digitization remains critical. Another challenge is trust. Researchers need assurance that AI isn’t “hallucinating.” The best AI-driven digital archives always link back to original sources, ensuring transparency.

The Road Ahead

Generative AI is still in its early stages for libraries, but the potential is enormous. Imagine querying, “What were the public health measures during cholera outbreaks in the 19th century?” Instead of a list of documents, the system delivers a synthesized narrative with citations. Or asking, “How did jazz spread through Europe in the 1920s?” and instantly seeing a cultural timeline.

This is not science fiction—it is already beginning.

From Preservation to Possibility

Digital libraries began as preservation projects. They are now evolving into intelligent systems that not only safeguard knowledge but amplify it. AI in digital libraries and AI-driven digital archives are not replacing researchers or librarians; they are empowering them.

At Ninestars, we believe this is the natural next step after digitization. Libraries and archives that embrace AI today will define how future generations interact with history, culture, and knowledge. It’s time to act on integrating AI into library services and reassert the role libraries have historically played in building future-ready knowledge economies.

Why OCR Accuracy Matters: The Cost of Mistakes

In the fast-paced digital world, where data is the backbone of decision-making, businesses increasingly rely on Optical Character Recognition (OCR) technology to process and extract information from vast amounts of documents. OCR is considered one of the key enablers of digital transformation, enabling organizations to convert physical documents into accessible digital data.

However, not all OCR solutions are created equal. While basic OCR systems can help read and extract text from scanned documents, their accuracy can vary widely. The OCR accuracy impacts the overall quality of extracted data ans processes that depend on it, and ultimately the business’s bottom line.

Inaccurate OCR = Business Risk

Inaccurate document processing leads to errors in data, causing operational disruptions, increased costs, and damage to a company’s reputation. OCR accuracy matters, and here’s why the cost of mistakes can be significant:

  1. Financial Implications of OCR Errors

For many businesses, OCR errors aren’t just an inconvenience—they can translate into direct financial losses. Most organizations rely on automation platforms that include OCR as a foundational component to process financial documents, invoices, and contracts. However, if the OCR component is inaccurate, it can create cascading errors throughout the automated workflow.

Invoice Errors: Consider a scenario where a finance team uses an Intelligent Document Processing (IDP) system to process invoices. If the OCR layer misreads an invoice total, payment terms, or vendor information, the company could accidentally overpay or underpay. Worse still, missing key fields like taxes or early payment discounts can delay processing and impact cash flow.

Contract Misinterpretation: In legal workflows, OCR is often responsible for the first step—digitizing and extracting key terms. If inaccuracies occur here, they can carry through contract review tools or compliance checks, leading to flawed interpretations, legal exposure, or missed deadlines.

Operational Costs: Poor OCR accuracy increases the need for manual review and correction downstream. Even in sophisticated IDP workflows, time and resources must be diverted to catch and fix mistakes. This reduces productivity and weakens the ROI on automation initiatives.

  1. Customer Experience at Risk

The accuracy of OCR within automation workflows directly impacts how customers experience your services. An error introduced by OCR early in the document lifecycle can ripple into customer-facing processes—leading to delays, incorrect communication, or billing issues.

Invoice and Billing Issues: Customers receiving invoices generated from inaccurate OCR outputs may find incorrect totals, missing details, or wrong references. While the system may automate document generation, the quality of that automation depends heavily on the OCR’s ability to extract data correctly in the first place.

Delayed Service or Errors in Orders: In industries like retail or logistics, OCR powers the initial intake of forms, order sheets, or shipment requests. If the OCR component misinterprets these documents, it can lead to downstream automation triggering incorrect actions—like sending the wrong items, scheduling delays, or duplicating orders.

A flawed OCR layer in your automation stack may be invisible to customers, but its effects certainly aren’t. Inaccuracies erode trust, delay service, and ultimately harm customer retention.

  1. Legal and Compliance Risks

In highly regulated industries such as finance, healthcare, and legal services, accuracy in document automation isn’t optional—it’s a matter of compliance. OCR plays a foundational role in these workflows, powering data extraction for systems that manage tax records, patient files, and contracts. If OCR introduces errors early in the automation pipeline, the consequences can be legally and financially severe.

Healthcare Compliance: In healthcare, OCR is used within automation platforms to extract patient data from forms, insurance documents, and medical records. Any error at the OCR stage can lead to incorrect or incomplete data flowing into electronic health record (EHR) systems. This could trigger HIPAA violations, impact patient care, or erode trust.

Financial Reporting: In the financial sector, OCR is often the first step in processing documents like tax returns, compliance filings, and audit reports. An inaccurate OCR output can corrupt downstream data analytics and reporting tools—leading to compliance breaches, audit flags, or regulatory penalties. In high-stakes environments, even a single field misread can cause substantial risk.

  1. Reduced Efficiency and Increased Error Propagation

OCR technology streamlines operations by reducing manual data entry. But when OCR accuracy is poor, it does the opposite—creating bottlenecks and increasing the likelihood of error propagation throughout your automated systems.

Manual Interventions: When an OCR engine misinterprets content, teams often have to manually verify and correct outputs within the broader automation flow. This manual intervention defeats the purpose of deploying automation in the first place and slows down processing times, reducing overall ROI.

Cascading Errors in Integrated Systems: Inaccurate OCR doesn’t just cause isolated issues—it affects every downstream system that relies on its output. For example, if OCR misreads a figure in an invoice, that faulty data could influence accounting entries, tax computations, and audit readiness. The more deeply integrated your systems are, the more widespread the impact of a single OCR error becomes.

  1. The Importance of Choosing an Accurate OCR Solution

To avoid the aforementioned risks, it’s crucial to choose an OCR solution that provides high levels of accuracy. While standard OCR technology can help with basic text recognition, it’s often limited in its capabilities to handle complex documents or ambiguous data. It’s vital to look for an OCR system that incorporates advanced AI and machine learning capabilities, like AOTM OCR, that can:

  • Adapt to Complex Documents: Recognize text in multi-page documents, complex layouts, and even handwritten notes.
  • Understand Context: Provide deeper contextual understanding to accurately extract and categorize data.
  • Automatically Correct Errors: Use AI to detect and correct errors in real-time, improving overall accuracy.
  • Process Multiple Languages: Offer multi-language support to extract data from documents in different languages with high precision.

By implementing an advanced OCR solution with AI-powered capabilities, businesses can ensure that their document processing is as accurate, efficient, and error-free as possible.

The Cost of Mistakes vs. The Value of Accuracy

OCR mistakes may seem minor at first, but their ripple effects can impact a business in many ways: from financial losses and customer dissatisfaction to legal liabilities and operational inefficiencies.

In today’s business environment, where data is gold, OCR is a critical component of automation and digital transformation. But the true value of OCR technology isn’t just in its ability to extract text—it’s in how accurately it does so. Choosing the right OCR system, like AOTM OCR, ensures that businesses extract, process, and utilize data with maximum precision, minimal errors, and greater efficiency.

Digital Preservation Challenges: Why METS and ALTO Are Essential for Large-Scale Archival Projects

Imagine discovering a centuries-old manuscript, brittle and broken at the edges. Now imagine being tasked to digitize and add it to your library’s collection—not just as a scanned image but as a fully searchable and preserved digital asset. That’s the kind of challenge libraries and archives worldwide face as they move from print to pixels.

These digital preservation initiatives are sometimes large scale, running into several years because it isn’t simply scanning. Without proper structuring, metadata, and text encoding, digital collections risk becoming unsearchable, unusable, or even obsolete. That’s where METS (Metadata Encoding & Transmission Standard) and ALTO (Analyzed Layout and Text Object) come in.

The Scale of the Challenge: Libraries as Data Giants
Major libraries and archives house millions—sometimes hundreds of millions—of items, with ongoing digitization efforts processing thousands of pages daily. Large-scale projects require more than high-resolution scans—they need interoperability, structured metadata, and full-text accuracy for users to meaningfully engage with the content.

Here’s the problem:

  • A simple image-based digital archive lacks context. A TIFF scan of a rare book is just a picture unless it’s properly indexed.
  • Poor OCR (Optical Character Recognition) results mean users can’t search the text accurately—especially in historical or non-Latin scripts.
  • If metadata isn’t standardized, collections become data silos, limiting interoperability across institutions.

METS: Bringing Structure to Digital Archives

METS is like a blueprint for digital objects. Instead of just storing a document, METS binds together multiple components—images, OCR text, metadata, and structural relationships—ensuring that a digitized book or newspaper is more than just a stack of files.

Why METS Matters:
Structural Mapping – Defines the order of pages, chapters, or multi-volume works.
Preservation Metadata – Ensures long-term digital viability by tracking technical details and provenance.
Interoperability – Enables seamless exchange across repositories (Europeana, HathiTrust, DPLA, ProQuest, JSTOR).

Think of METS as a librarian’s guide for the digital world—a way to organize and ensure long-term usability of complex digitized collections.

Why ALTO is the Unsung Hero of Searchability

OCR alone isn’t enough. Standard OCR might extract text, but it loses layout details—crucial for newspapers, tables, and manuscripts. ALTO fixes that.

What ALTO Does Differently:

  • Retains Text Layout – Captures columns, footnotes, and even marginalia, making digitized newspapers or periodicals look like their physical counterparts.
  • Improves Search Accuracy – Maps text positions to original layouts, reducing OCR errors.
  • Supports Multilingual & Historical Texts – Handles complex scripts, Fraktur fonts, and even handwritten materials.

Example: As the Exclusive Partner for ProQuest’s Historical Newspapers Program (HNP) since 2001, we have digitized iconic publications like The New York Times, The Wall Street Journal, and many more. Using METS/ALTO, we have structured 28 million pages across 55 newspaper titles—some dating back to 1764—ensuring that every article, photograph, and advertisement is fully searchable and meticulously preserved. Our solutions not only safeguard history but also create revenue opportunities through content distribution and digital accessibility across tablets, smartphones, and emerging platforms.

Fun Fact: Digital Archives Are at Risk—Even Digital Ones!

Did you know that NASA lost the original high-resolution recordings of the 1969 moon landing? The tapes were overwritten due to poor archival practices. Digital doesn’t always mean permanent—without proper structuring (like METS/ALTO), even digital archives can disappear over time.

Future-Proofing Archives with METS & ALTO

In an era where digital libraries are growing exponentially, METS and ALTO are non-negotiable. They make sure that today’s digitization efforts remain accessible and meaningful for decades—even centuries—to come.

For libraries, archives, and cultural institutions, the choice is clear: Digitize, but do it right.

Ninestars: Bringing Structure to Digital Archives

At Ninestars, we go beyond digitization—we ensure archives are structured, searchable, and future-proof. Our solutions include AOTM OCR, indexing, and metadata enrichment to enhance content discoverability.

With METS, ALTO, MARC, and Dublin Core-compliant workflows, we’ve digitized 1.2 billion pages to date, making vast collections accessible across libraries, enterprises, and institutions. Our expertise spans subject- and keyword-based indexing, AI-powered OCR for handwritten texts, and contextual OCR in 71 languages for unmatched accuracy.

From national archives to rare manuscripts, we help organizations preserve history while unlocking new revenue and digital opportunities. Let’s talk.

What We Learned at WAN2025: AI and the Future of Newsrooms

The World News Media Congress 2025 (WNMC25) in Kraków has officially concluded, leaving behind a wealth of insights that continue to shape how we view the intersection of journalism and technology. As proud sponsors of the event, Ninestars had the opportunity to engage with the brightest minds in media and technology, gaining invaluable perspectives that are driving the future of news.

The Congress highlighted a new wave of media transformation, driven by technological innovation, AI integration, and a renewed focus on providing real value to audiences. These advances are not just improving the quality and efficiency of news production but are also setting the stage for a media landscape where personalization, audience engagement, and ethical AI take centre stage.

AI and the Transformation of Newsrooms

A big theme of WNMC25 was the integration of AI in journalism, an undeniable trend that has moved beyond speculation and into action. AI is no longer a buzzword or a distant possibility; it is being embedded in the day-to-day operations of newsrooms worldwide. From editorial workflows to content creation, AI is playing an increasingly pivotal role in how stories are told and consumed.

One of the most profound insights from the Congress was the increasing reliance on Generative AI. Speakers shared real-world examples of how this technology is already streamlining content creation, improving productivity, and expanding audience reach. AI tools are now integral in supporting editorial decisions, from helping journalists gather data to automating repetitive tasks. The focus is clear: AI must be implemented in a way that enhances editorial workflows and maintains the values of trust and accuracy, which are the bedrock of quality journalism.

At Ninestars, we’re proud to align with this vision. Our AOTM Intelligent Automation Platform is designed to empower newsrooms with the speed and precision they need to process vast volumes of content. With AOTM OCR (Optical Character Recognition) and AOTM ICP (Intelligent Content Processing), we’re helping newsrooms handle information faster and more accurately, which ultimately allows them to focus on what matters: producing high-quality journalism.

AI’s Role in Personalized Journalism

Personalization is no longer just a luxury for newsrooms; it’s a necessity. As AI continues to evolve, it provides new opportunities to tailor content to the specific preferences and behaviours of individual readers. During the congress, the idea of audience-centric strategies was discussed in depth. News organizations are increasingly leveraging AI to deliver personalized experiences that engage readers at a deeper level. This means not just creating content that is relevant, but making sure it resonates at a personal level.

For example, AI-driven personalization is allowing publishers to adjust the content they provide based on data, whether it’s user behaviour, geographic location, or even social trends. Short-form content is also becoming more influential in reaching younger audiences, especially Gen Z, who demand quick, digestible news that fits into their daily lives.

Ninestars is fully committed to empowering publishers with these AI-driven personalization strategies. Our solutions help streamline content processing, automate repetitive tasks, and deliver deep insights that make it easier to engage audiences in meaningful ways.

Ethics, Trust, and the Future of Journalism

The conversations at WNMC25 weren’t just about technology; they also focused on the broader ethical implications of AI in journalism. As AI becomes more ingrained in newsrooms, ensuring that it supports the values of trust, transparency, and editorial independence is crucial. The term Authentic Intelligence emerged as a key theme, emphasizing the need for AI to be used responsibly in ways that bolster the integrity of journalism rather than undermine it.

Industry leaders like Ingrid Verschuren from Dow Jones and Tom Rubin from OpenAI highlighted the importance of grounding AI in strong ethical frameworks. They stressed that AI should empower journalists, not replace them, and that AI systems should be transparent, accountable, and aligned with the values of responsible journalism. These conversations were important in reminding us that as AI becomes more advanced, we must be vigilant in maintaining the trust of our audience.

At Ninestars, we are committed to developing AI solutions that respect these ethical considerations. Our platform is designed to automate and streamline processes while upholding the principles that make journalism a trusted source of information and perspectives. From responsible data usage to transparency in AI decision-making, we ensure that our technology supports the greater good of the industry.

Looking Ahead: A Smarter, More Efficient Future

As WNMC25 wrapped up, the focus was clear: The future of journalism will be defined by AI, but it’s how we use it that will determine its impact. AI is not just about efficiency; it’s about improving quality, enhancing the audience experience, and enabling news organizations to focus on what they do best: telling great stories.

As Ninestars continues to work alongside media companies, we are proud to be part of this transformation. We are actively building solutions that not only help publishers streamline their workflows but also foster stronger connections with their readers. The future of media is bright, and with AI as an enabler, newsrooms can rise to the challenge of staying relevant in an increasingly digital world.

The World News Media Congress 2025 was a powerful reminder of the importance of AI in shaping the future of journalism. From enhancing editorial workflows to creating personalized experiences, AI is helping newsrooms embrace the future while staying true to their core values. As the event concluded, it was clear that the momentum toward AI-driven innovation in media is only going to grow stronger.

We’re excited to continue our journey with the media industry, working hand-in-hand with publishers to build a smarter, more efficient future for journalism. Thank you to everyone who shared their insights and helped shape these important conversations. The journey has just begun, and at Ninestars, we are ready to continue making an impact.

TL;DR

Key Insights from WNMC 2025

  • Generative AI is revolutionizing content creation, enabling newsrooms to streamline processes, boost productivity, and improve engagement.
  • Personalized journalism is now a strategic necessity, with AI allowing publishers to create tailored content that resonates with individual audiences.
  • Ethical AI remains a focal point, with leaders emphasizing the need for AI to enhance, rather than replace, journalistic integrity and trust.
  • AI is already transforming newsrooms by enhancing editorial workflows and content creation.

Discover how Ninestars is helping newsrooms thrive in the digital age: Explore here

Ninestars at the World News Media Congress 2025: Shaping the Future of Journalism with AI-Powered Innovation

We are excited to announce that Ninestars Information Technologies Pvt. Ltd. will be participating in the World News Media Congress 2025, happening in Krakow, Poland from May 4-6, 2025! This premier event brings together the brightest minds from across the global media industry to explore new strategies, innovations, and solutions in journalism. As we prepare to showcase our AI-driven solutions at the Congress, we look forward to demonstrating how Ninestars is revolutionizing the future of newsrooms.

Why Ninestars is Here

The media landscape is evolving rapidly, and Ninestars is at the forefront of this transformation. With over 26 years of expertise in the media and publishing sector, we have always been committed to empowering organizations with intelligent, scalable, and future-proof solutions. At the World News Media Congress 2025, we aim to highlight how our AI-powered platforms and R&D capabilities are helping newsrooms stay ahead of the curve.

What We’re Showcasing

At our booth, we will showcase a comprehensive range of AI-driven solutions designed to address the unique challenges of modern journalism. Our offerings include:

AOTM OCR: Advanced AI-powered optical character recognition for transforming printed media into actionable data.

AOTM GPT: A generative AI engine tailored specifically for newsrooms, helping to accelerate content creation while maintaining editorial integrity.

AOTM ICP: Our Intelligent Content Processing platform that intelligently ingests, indexes, and processes diverse content types.

Archive Transformation and Monetization: Turning valuable archives into dynamic assets that drive contemporary storytelling and revenue generation.

Hyper-Personalized News: AI-powered personalization engines to tailor content and advertisements based on reader behavior.

Additionally, we will demonstrate our R&D capabilities, including advancements in Editorial AI Models, DSLMs, LLMs (Large Language Models), and Multimodal AI, as well as Computer Vision for media content analysis and enhancement.

What to Expect at the Congress

The World News Media Congress will feature three dynamic summits, deep-dive sessions into future media trends, and an exhibition space where industry leaders will share insights on technology and innovation. With social events and networking opportunities, it’s a perfect setting to engage with fellow professionals in the media and publishing world.

Ninestars is proud to be part of this exciting event, and we are eager to share how our AI-driven solutions can help news organizations optimize their workflows, create better content, and unlock new business opportunities.

Connect with Us

We invite you to visit our booth and engage with our experts as we demonstrate our latest solutions. Let’s discuss how we can help your newsroom stay ahead with AI-powered innovation.

We are excited to connect with industry leaders, journalists, and innovators in Krakow and contribute to the ongoing evolution of media.

Stay tuned for more updates as we approach the World News Media Congress 2025. We look forward to seeing you there!

Explore how we can help you. Learn more here:
https://wan2025.ninestarsglobal.com

The Future of News: How Technology Will Define the Next Chapter for Publishers

As we prepare to join the conversation at World News Media Congress 2025 at Krakow next week, we dove deep into two reports setting the agenda for global publishing — Innovation in News Media World Report 2024–2025 by Innovation Media Consulting Group and WAN-IFRA’s own World Press Trends Outlook 2024–2025.

While these reports capture industry-wide shifts, we examined them through a sharper lens—the accelerating role of technology and AI in defining the next era of journalism. 

Here’s our technology-first take on what it means and where publishers must go next.

The Industry Shift: Print Is Shrinking, Trust Still Matters 

Global newspaper circulation has halved over the past decade. Ad revenue continues migrating to digital giants. Yet one constant remains: audiences still crave credible, local, in-depth journalism — they’re simply consuming it differently. 

As platforms like Grok, DeepSeek, and a growing roster of AI disruptors reshape information ecosystems, publishers must ask:
If the audience has changed, why haven’t we? 

Strategy 1: From Print Products to Platform Ecosystems 

Leading publishers are moving from newspaper-as-product to news-as-a-service. They are building interconnected ecosystems—responsive websites, audio integration, video explainers, newsletters, podcasts—all feeding into personalized user journeys. 

Case in point: The South China Morning Post revamped itself into a subscription-first digital platform, achieving a 32% YoY growth in digital subscriptions in early 2025. 

At the heart of this evolution? Data, personalization, and multi-format storytelling. 

Strategy 2: Intelligent Archiving: Unlocking New Revenue Streams 

Newsrooms sit on decades, sometimes centuries, of invaluable content. AI-led digitization is transforming archives into dynamic, monetizable assets.

We see publishers adopting: 

  • OCR, NLP, and AI tagging to make archives searchable and accessible 
  • New subscription products for researchers, schools, and history enthusiasts 
  • Repurposed archival content for documentaries, timelines, and “On This Day” features 

Insight: Digitized archives boosted SEO traffic by 15–20% between 2020–2024, and this is just the beginning. 

Strategy 3: AI for Newsroom Efficiency and Personalization 

AI is no longer theoretical — it’s practical, operational, and indispensable.
Smart newsrooms in 2025 are deploying AI for: 

  • Automated tagging and summarization to accelerate publishing 
  • Predictive analytics to deliver more relevant, engaging stories 
  • AI-assisted live reporting for financial results, elections, and sports 
  • Multilingual content generation to expand global reach 

In a landscape dominated by rapid news cycles and AI aggregators, intelligent automation isn’t a luxury — it’s survival. 

Strategy 4: Going Hyperlocal and Winning 

While global news is increasingly commoditized, trust and proximity are premium currencies.

Savvy publishers are: 

  • Launching hyperlocal editions for districts, cities, and even neighbourhoods 
  • Partnering with libraries and NGOs to syndicate trusted content 
  • Building exclusive investigative verticals with subscription access 

Trend to watch: In several markets, local news subscriptions are now outpacing national ones. 

Strategy 5: New Monetization Models: Beyond Ads 

The ad-supported model is losing steam but reader-supported journalism is accelerating. 

Publishers are exploring: 

  • Metered paywalls balanced with sampling strategies 
  • Micro-payments for one-off article access 
  • Reader memberships with value-adds like exclusive newsletters, Q&As, and events 
  • Strategic bundling with OTT, e-learning, or other services 

Key lesson: In 2025, trustworthy content is a premium experience and audiences are willing to pay for it. 

Strategy 6: Talent and Infrastructure for a Digital-First Future 

Tech adoption alone isn’t enough.
Publishers must invest in people and platforms simultaneously. 

This means: 

  • Upskilling journalists in multimedia storytelling, analytics, and AI tools 
  • Hiring data journalists, podcast producers, and UX designers 
  • Building CMS platforms that are content-creation, distribution, and analytics hubs all-in-one 

The Bigger Payoff: Reclaiming Influence 

Digitized, intelligent newsrooms are not just surviving; they are redefining their societal role: 

  • Guardians of truth in a fragmented information economy 
  • Community anchors spotlighting hyperlocal issues 
  • Multimedia educators driving informed citizenship 

Future-Proofing with Intelligent Automation 

At Ninestars, we don’t just digitize — we help news organizations reimagine journalism itself.

From custom AI R&D that shapes newsroom-first solutions, to generative engines built with an editorial backbone, to platforms that turn archives into living assets, we bring technology, language intelligence, and audience insights together to fuel sustainable growth.

Whether it’s empowering investigative depth, driving hyper-personalization, or unlocking new revenue streams, we build the invisible infrastructure that future-proofs newsrooms for the AI era — with context, credibility, and creativity at the core.

Future-proof your newsroom with AI-powered solutions— reach out to our experts today.  

AOTM OCR vs. Traditional OCR: A Head-to-Head Comparison

OCR is the silent magic behind digitizing documents, but traditional OCR has its limits. Enter AOTM OCR—AI-powered, multilingual, and built for complex layouts. From blurry scans to handwritten text, AOTM OCR ensures precision where traditional OCR stumbles. Smarter, faster, and adaptable, it’s the future of document processing.

There are some technical terms we casually drop in conversations or project discussions without fully appreciating the brilliance behind them— Optical Character Recognition (OCR) is one such term. It might sound like a technical jargon that only tech enthusiasts or data processing experts throw around but OCR is, in fact, the silent magic behind numerous activities, like scanning receipts, digitizing analogue archives, or even auto-filling information on forms.

Think of OCR as the unsung hero, the bridge that connects physical ink on paper to the digital realm. OCR converts the static, inaccessible printed assets into an editable, searchable digital format. With OCR, content in analogue formats come to life as accessible, searchable and editable assets, perfectly aligned with today’s digital world.

The origins of OCR trace back to the late 1920s—before modern computers were even a concept! In 1929, German engineer Gustav Tauschek developed the first OCR machine. While its capabilities were limited, this invention set the stage for a digitization revolution that would follow decades later. Here’s a fun tidbit: OCR technology played a role during World War II, assisting blind veterans in reading their mail. Ray Kurzweil’s innovations in OCR, especially those aimed at reading text aloud, were initially created to support the visually impaired.

The journey of OCR: From mechanical eyes to AI-powered engines

 The story of OCR’s evolution is nothing short of fascinating. In the 1950s used by institutions like the U.S. Postal Service and IBM for automated mail sorting and check processing, OCR was a mechanical innovation. In the 1970s, Ray Kurzweil, a futurist and inventor, created the first omni-font OCR system, which could read text in any typeface. This was a major breakthrough!

Over the decades, OCR technology steadily improved, driven by innovators and major tech players. Companies like ABBYY, Adobe, and Google have been leading the charge, turning OCR from a niche technology into a widespread tool used in banking, healthcare, law, education, etc. Today, tools like ABBYY FineReader and Tesseract are everyday staples in content digitization.

But as remarkable as traditional OCR has been, new technologies are pushing the boundaries of what’s possible. Enter AOTM OCR, the AI-powered OCR that is redefining document recognition.

AOTM OCR vs. Traditional OCR: What’s the Difference?

The key difference between traditional OCR and AOTM OCR lies in the integration of artificial intelligence and machine learning, making AOTM OCR a game-changer especially when extracting data from low-quality or damaged documents. But let’s break down their differences in a head-to-head comparison:

Traditional OCR: Tried, Tested, But Limited

Traditional OCR has been reliable for years, especially for digitizing books, simple forms, and converting typed or printed documents into searchable formats. However, it has some limitations:

  • Accuracy issues: When handling complex documents, handwritten texts, or blurry fonts, traditional OCR struggles to maintain high accuracy.
  • Limited language support: While it works well with Latin-based languages, it often falters with scripts like non-Latin characters or Indic languages.
  • Rigid data extraction: Traditional OCR systems are relatively inflexible, making it difficult to accurately extract complex data like tables or structured fields.
  • Inconsistent table recognition: Extracting content from tables or structured data is a challenge, often leading to inaccuracies.

AOTM OCR: AI-Powered Document Processing

AOTM OCR uses artificial intelligence and machine learning to enhance accuracy and adaptability. Here’s how AOTM OCR stands out:

  • Multi-language mastery: AOTM OCR supports 70+ languages, including Indic languages. This makes it a versatile tool for global companies dealing with multi-lingual documentation.
  • Holistic detection strategy: AOTM OCR doesn’t follow a one-size-fits-all approach. Its AI-powered holistic detection adapts to specific industries—whether it’s healthcare, finance, or legal—ensuring accurate data extraction tailored to the domain.
  • Partial character detection and auto-correction: In older or damaged documents, some characters may be smudged or incomplete. While traditional OCR systems often fail to recognize these, AOTM OCR’s AI engine intelligently predicts and fills in missing characters, providing much higher accuracy.
  • Advanced table detection and content segmentation: AOTM OCR excels with advanced algorithms designed to detect and segment content accurately. Whether it’s legal documents, medical records, or financial reports, AOTM OCR ensures precision where traditional OCR stumbles.
  • Robust segmentation and AI recognition: Powered by AI, AOTM OCR excels in recognizing text across diverse formats, even with complex fonts, unstructured layouts, or scanned documents with mixed content. The system is built to handle what traditional OCR often can’t.

Traditional OCR: still relevant but lagging behind

To give traditional OCR its due credit, it’s still an efficient tool. Here’s where it continues to perform well:

  • Basic text recognition: Traditional OCR handles clean, typed documents fairly well, making it a good option for scanning books or printed invoices.
  • Cost-effective for basic needs: If your document processing needs are basic and don’t require complex extractions, traditional OCR remains an affordable option.

But when it comes to more complex scenarios—think handling handwritten forms with varying legibility, processing documents that feature a mix of fonts and styles, or tackling multi-lingual texts—traditional OCR begins to falter. This is especially true in specific domains, such as the complexities of legal documents with diverse layouts, the multilingual nature of international contracts, etc., where precision and adaptability are crucial. In contrast, AOTM OCR is built to thrive in these challenging environments.

AOTM OCR vs. Traditional OCR: A Feature Comparison

Feature AOTM OCR Traditional OCR
Accuracy Superior AI-powered precision Decent but struggles with complexity, especially in low-quality documents
Language Support 70+ languages including Indic Largely limited to Latin-based languages
Table Detection Advanced and accurate Inconsistent
Partial Character Detection AI-driven, auto-correction Often misses or misreads characters
Domain-Specific Customization Tailored to industries like healthcare, finance, etc. Generic, not domain-specific
Deployment SDK, Cloud SaaS, API Limited to standalone installation

AOTM OCR is the Future of Document Processing

As businesses move toward more complex, data-driven operations, the limitations of traditional OCR are becoming clear. While traditional OCR still holds value for basic tasks, AOTM OCR offers the advanced AI-powered capabilities that modern enterprises need.

For those wanting unparalleled efficiency and accuracy in their document workflows, AOTM OCR represents the next big leap in OCR technology, outclassing its traditional counterparts and setting a new standard for document processing.

We hope this information has sparked your interest in the potential of AOTM OCR. If you’re ready to enhance your document processing, reach out to us at Ninestars. Let’s explore how AOTM OCR can make a difference for your business!