AI - NewForge Tech

Record Upload & Transform: How to Capture Perfect Videos For AI-Powered Work Instructions

Javier Ortiz — Thu, 15 Aug 2024 14:37:29 +0000

Hey there, NEW FORGERS!

So, you’ve got a process to document, a person retiring that needs a brain dump, or a need to create a troubleshooting guide, and you’re ready to dive into the world of recording NewForge Video Instructions and AI video narration. Before you hit (or speak if you are using smartglasses) that record button, let’s talk about how to make your videos not just good, but epic.

But wait- have you heard about our new Video Instructions feature? If not, let me fill you in! This handy tool lets you record work instructions, processes, or procedures using smartglasses, tablets, or even your mobile device. Once recorded, your video gets automatically uploaded and ingested by our AI. From there, you can ask the AI to generate detailed work instructions, create learning materials, or even answer troubleshooting questions based on your video.

The best part? Your original video, transcription, timestamps, and any documentation generated are securely stored in your enterprise library.

Check out this video of our client, B/S/H, recording their processes using smartglasses!

Of course, the magic only works as well as the content you provide. That’s why we’ve put together these best practices to ensure your videos are as awesome and useful as they can be. And don’t worry, we won’t let you go it alone—consider us your trusty sidekick in this journey (but without the tights).

1. Set up equipment or smartglasses

If you are using smartglasses, make sure they are connected to Wi-Fi
If you are using mobile devices or tablets, it is recommended you use a tripod and external microphone

2. Pre-Recording Checklist

Prepare your script or outline
Gather all necessary tools and equipment
Set up your workspace
Do a test recording to check audio and video quality

3. Introduction

State your name
Mention the equipment or procedure you’re narrating – include as much detail as needed to identify the equipment or procedure (eg. Model number, serial number, workcenter, line number)

Make sure to prep before recording!

4. Content Structure

Give an overview of the process
List required tools and materials
List required PPE, safety precautions or concerns, LOTO requirements
Explain each step in a logical order
Highlight important safety precautions
Mention common pitfalls or mistakes to avoid

5. Narration Best Practices

Speak clearly and at a moderate pace
Use simple, concise language
Explain technical terms when first introduced
Describe what you’re doing as you’re doing it
Mention visual cues (e.g., “the red wire”, “the square button”)

6. Visual Best Practices

Show close-ups of intricate parts or actions
Demonstrate correct tool usage
Use on-screen text for part names or measurements
Ensure your hands don’t obstruct the view

7. Conclusion

Summarize key points
Mention any final testing or quality checks
Provide resources for further information or support

Always review your content before uploading!

8. Post Recording

Review the video for clarity and completeness
Edit if necessary, adding captions or annotations

Why This Information Is Important:

Consistency: Following these guidelines ensures all videos maintain a standard format, making them easier to follow and use.
Clarity: Providing your name and specific equipment details helps viewers quickly identify the relevant content.
Comprehensiveness: Covering all steps thoroughly aids in accurate reproduction of the process.
Safety: Highlighting precautions can prevent accidents or equipment damage.
AI Integration: Clear, structured narrations improve transcription accuracy and make the content more useful for AI-based knowledge retrieval systems.
Future-Proofing: Detailed explanations and visual cues help preserve knowledge that might otherwise be lost over time.

Remember: These narrations will be transcribed and used for AI prompting, so clarity, accuracy, and thoroughness are crucial for creating valuable training data.

And that’s it! Following these best practices isn’t just about looking and sounding like a pro—it’s about creating clear, accurate, and useful work instructions that stand the test of time.

If you hit a snag or just want to bounce ideas off someone, we’re here to help. Drop us a line, and let’s make sure your Video Instruction is one for the books!

The post Record Upload & Transform: How to Capture Perfect Videos For AI-Powered Work Instructions first appeared on NewForge Tech.

Top 10 AI Terms You Need to Know

Emily Hildebrand — Tue, 18 Jun 2024 18:12:24 +0000

AI has hit the industry HARD and staying up to date on the long list of terms people throw around when they talk about it can be quite a challenge. That’s why we’ve compiled the top 10 AI terms you should know to be “in the know.”

Artificial Intelligence (AI): Let’s start with the basics, AI is a broad term for mechanical systems that act like humans. AI’s learn from different techniques that we’ll discuss later in this article like Machine Learning and Deep Learning.

Check out this article to learn more about AI and its use cases in manufacturing.

Machine Learning (ML): According to IBM, “Machine learning (ML) is a branch of artificial intelligence (AI) and computer science that focuses on using data and algorithms to enable AI to imitate the way that humans learn, gradually improving its accuracy.” In simpler terms, it’s the process of feeding a computer a large amount of data that it uses to learn from with time.

For example, a manufacturer may use ML in predictive maintenance. An ML algorithm could analyze data from sensors on manufacturing equipment to predict when the machine may go down.

Check out our video to hear our co-founder’s take on AI vs ML.

Deep Learning: Deep Learning is a more advanced type of machine learning that uses many layers of processing units to analyze and understand complex data. It’s inspired by how our brain works, with each layer of “neurons” learning to recognize different features. This is how technologies like facial recognition in social media apps and voice assistants work so well.

Artificial Neural Networks : Neural Networks are the backbone of deep learning. They are a series of algorithms that try to recognize patterns, similar to the way a human brain does.

A neural network consists of interconnected layers of nodes (or neurons) that process data. Think of it like a giant web of decision points that help a computer learn and make sense of data, such as recognizing handwriting.

Natural Language Processing (NLP): NLP is a field of AI that helps computers understand, interpret, and respond to human language. It’s the technology behind things like chatbots, translation services, and voice-activated assistants.

When you talk to your phone or type in a search query, NLP helps the computer understand what you mean and respond appropriately.

OpenAI’s ChatGPT & NewForge’s ExpertAI both use NLP to allow users to use their natural word-choice to communicate with their AI’s.

Computer Vision: Computer Vision is a type of AI that enables computers to interpret and understand visual information from the world, like images and videos. It’s like giving sight to machines.

For example, it can be used for bin picking, palletizing and depalletizing, machine tending, defect detection, & much more.

Supervised Learning: Supervised Learning is a machine learning technique where a computer is trained on a labeled dataset, meaning the data includes both the input and the correct output.

For example, if you wanted to teach a computer to recognize cats in photos, you’d show it thousands of photos labeled as “cat” or “not cat” until it learns to identify them on its own.

People are starting to use supervised learning for things like spam detection and predictive analytics.

Unsupervised Learning: Unsupervised Learning is another machine learning technique where the computer is given data without labels and must find patterns and relationships on its own. It’s like exploring a new city without a map and discovering landmarks and shortcuts on your own.

This method is used for things like organizing large sets of images by similarity or finding customer segments in marketing data.

Reinforcement Learning: Reinforcement Learning is a type of machine learning where an agent learns to make decisions by taking actions and receiving rewards or penalties. It’s similar to training a pet with rewards for good behavior and penalties for bad behavior.

This approach is used in robotics for teaching robots to perform tasks, in video games for developing intelligent opponents, and in self-driving cars for navigating roads safely.

Training Data: Training Data is the dataset used to teach an AI model how to make decisions or predictions. It includes inputs and the correct outputs so the model can learn their relationship.

For example, to train a model to recognize handwritten digits, you’d use a large collection of images of digits along with their correct labels.

Understanding these fundamental AI terms is crucial for anyone looking to stay current in the rapidly evolving field of artificial intelligence. From the basics of AI and Machine Learning to the intricacies of Deep Learning and Natural Language Processing, each concept plays a significant role in shaping the technologies that are transforming industries today.

As AI continues to advance, keeping on top of of these key concepts will be essential for leveraging its full potential and staying ahead in a tech-driven world.

The post Top 10 AI Terms You Need to Know first appeared on NewForge Tech.

Artificial Intelligence: New Solution to an Old Problem

Devki Sharma — Sun, 31 Mar 2024 09:43:21 +0000

In manufacturing and equipment-run industries, one issue is always going to be reoccurring, machine downtime. It’s unavoidable and the longer machines are down, the more money is lost. A lack of time and expert technicians only makes the problem worse.

NewForge has been focused on helping companies address this problem for years now. Their new platform, Knowledge Place, is now using Artificial Intelligence (AI) to bring companies the vital knowledge they need to get machines running as soon as possible.

What is Artificial Intelligence (AI)?

If you haven’t heard about AI, the concept can be simplified into mechanical “systems that act like humans,” according to IBM.

Artificial Intelligence has been around for a fairly long time. Conceptually, there have always been stories about artificial beings or machines that recreate human intelligence. Think of HAL 9000 in Stanley Kubrik’s 2001: A Space Odyssey or even farther back, Mary Shelley’s Frankenstein.

However, the term was first introduced in 1956 by John McCarthy during the Dartmouth Workshop where the goal was to make machines capable of human intelligence. John McCarthy has since been coined the “Father of Artificial Intelligence”.

But it wasn’t until around January 2023 that AI surged in popularity, with OpenAI’s introduction of ChatGPT-3. ChatGPT is a language processing tool that chats with a user. Unlike other chatbots, ChatGPT can remember and refer to things previously discussed in the conversation, simulating human-like interactions.

Overnight, ChatGPT became a household name. People found that it could be used to not only answer questions but also assist with tasks like writing emails, code, and essays.

NewForge Prior to AI

Before the introduction of AI into their solution, NewForge’s focus was creating custom augmented reality (AR) solutions for manufacturing and logistics that increase efficiency for their clients.

Here’s a simplified example of how their previous solutions worked: On the plant floor, a technician would wear a pair of smart glasses (picture a tablet that sits next to your eye) and receive all the instructions on how to do their task right on the screen. All they have to do is say out loud, “Continue” and the glasses will take them to the next step in their process.

If something went wrong, the technician could also call for help using their smart glasses. Through the camera, someone could see what they’re doing almost like they’re standing over their shoulder.

This solution helps frontline workers get their work done faster and reduce downtime all while reducing error rates.

Even though NewForge’s AR solutions have a surplus of benefits, NewForge noticed that some companies struggled with creating the instructions that go on the smart glasses.

ExpertAI: Say Hello to Instant Work Instructions

“It’s a massive pain in the ass,” Javier Ortiz, CEO and Co-founder of NewForge, said. “Nobody wants to recreate work instructions.”

“If you ask somebody that’s on the manufacturing floor… if you could have anything when it comes to work instructions, what would you want?”

People would tell you they want them to instantly POOF be created, Ortiz explained. “So that’s basically what we’re adding,” Ortiz said, referring to their new AI feature, ExpertAI.

ExpertAI is a new feature for NewForge’s comprehensive knowledge-sharing platform, NewForge Knowledge Place.

The generative AI tool not only provides quick troubleshooting answers but also creates work instructions for manufacturers. It eliminates the need for someone to take hours out of their day to recreate their processes into AR work instructions.

ExpertAI is one-way users can get instant help on Knowledge Place, but not the only way. Knowledge Place also provides a repository of machinery experts that users can schedule an appointment with and get instant help right when they need it.

How does ExpertAI work?

In simple terms, ExpertAI works similarly to ChatGPT. Except all its knowledge is surrounding equipment maintenance. A user simply writes in a question or request and the system responds.

For example, you could ask ExpertAI to help you troubleshoot your CNC machine or prompt it to create inspection instructions you could follow on your phone or smart glasses.

People usually have two main questions about ExpertAI: Where does the information come from? How do you keep things accurate?

To answer both, ExpertAI gathers information and learns from its users. Through NewForge’s “Feed the AI” feature, professionals can insert manuals, pictures, or any other form of media that the AI can then use to learn from.

One extremely useful feature is that ExpertAI can also learn from support calls. We have customers with all these previously recorded live calls that have the answers in them that are just going to waste, Ortiz said. Recorded calls are simply transcribed into written words that the AI can pull information from.

Users always have the option to give ExpertAI feedback on the responses it gives them.

For example, if a user were to prompt it for a set of instructions for calibrating a machine and step 4 wasn’t right, the user can let ExpertAI know to adjust that step.

From then on, the system knows that it needs to improve in that area and works to solve the issue or alert staff to let them adjust.

Is AI Taking my Job?

The short answer, No.

When looking at ExpertAI in the manufacturing, logistics, and equipment maintenance fields, it addresses gaps that desperately need to be filled. Not to take away work that’s going to be missed.

In these industries, there are just not enough people with the time or skills to create instructions, organize manuals, or do any of the other tasks that are essentially just to give other people the knowledge they need.

In addition, enterprises that use Knowledge Place (and ExpertAI in particular) for customer success, have more time to help more customers. Instead of spending an hour on a call for a simple question, customers can get instant feedback from the AI, also increasing their satisfaction rates.

When it comes to machine maintenance and repairs, sometimes, you just need to talk to another person. NewForge looks to use ExpertAI as its first line of defense against equipment downtime, not its only line of defense.

Knowledge Place also allows you to find experts in specific machinery and book an appointment to get help. These experts are then paid for their time and effort through the platform.

NewForge Knowledge Place is designed to be the knowledge tool that has all the help you need in one place.

What’s the Future of AI & ExpertAI?

The future of AI is dynamic with promises of changes in our society, economy, and technology. As it continues to find a place in our society, it’s important to find ways that it can be used to solve long-term issues and pain points.

NewForge has developed a way to implement AI in industries that are in desperate need of assistance.

As it currently stands, ExpertAI is in the development stage on Knowledge Place, but it should be ready for the public in early March according to Ortiz.

In the meantime, other features are ready to be used on the Beta of Knowledge Place. NewForge is currently seeking new users to join the forefront of the maintenance revolution.

To learn more about NewForge Knowledge Place or its ExpertAI feature check out the site or reach out to the NewForge team at [email protected]

The post Artificial Intelligence: New Solution to an Old Problem first appeared on NewForge Tech.