If you’ve ever needed to translate and create captions from a Swahili interview—or any other language not natively supported in Adobe Premiere—you’ll be pleased to know that it’s easier than you might think, thanks to AI technology. Here’s a step-by-step guide to help you through the process.

I created this guide for the students in our upcoming Storytellers Abroad Multimedia Workshop, traveling to Dar es Salaam, Tanzania, in a few weeks. These students will be capturing powerful stories from local nationals who primarily speak Swahili. Given the language barrier, I wanted to equip them with the tools and knowledge to effectively translate and create English captions for their interviews, ensuring that a global audience can understand and appreciate the voices and stories they document.

Step 1: Set Up Your Amazon Web Services (AWS) Account

You’ll first need an account with Amazon Web Services (AWS), which offers powerful translation tools. Setting up an account is straightforward:

1. Create an AWS Account: Go to the AWS website and follow the simple instructions to create a free account. AWS supports translation for 75 languages, making it a versatile tool for many projects beyond just Swahili.

Step 2: Create a Free ChatGPT Account

Next, you’ll want to use ChatGPT, an AI tool that can assist you in refining translations and creating captions. Follow these steps to get started:

1. Visit the ChatGPT Website: Click here to navigate to the ChatGPT site.
2. Sign Up: On the homepage, click the “Sign Up” button to create your account.
3. Verify Your Email: You’ll receive an email verification link after signing up. Click it to confirm your account.
4. Configure Your Account: Complete any additional account setup steps as prompted.
5. Explore Subscription Plans (Optional): While a free account is sufficient for most tasks, there are paid options if you require advanced features.
6. Test ChatGPT: Once your account is set up, you can use ChatGPT to test and refine your translations.

Step 3: Translate and Caption Your Video

Now that your accounts are set up, you can translate and caption your video. Here’s how:

1. Download Swahili Video: Video to use to practice
2. Use Adobe Premiere: Edit the video so that the voice is clear. I recommend using “Enhance” in the Audio Panel. Export the video or just the soundtrack.
3. Use AWS for Translation: Upload the Swahili audio to AWS’s translation service, selecting Swahili as the target language.
4. Refine with ChatGPT: Once you have a basic translation, use ChatGPT to translate the SRT Swahili into English SRT. I recommend, “Translate Swahili into English keeping SRT format exactly.”
5. Create Captions: After finalizing the translation, you can generate captions that are easily integrated into video editing software like Adobe Premiere.

This is a video walking you through all the steps

Expanding to Other Languages

The process outlined here isn’t just limited to Swahili. With AWS supporting 75 languages, you can apply these steps to translate and caption interviews in a wide range of languages, even those not directly supported by Adobe Premiere.

This method allows you to bridge language gaps in your video content, making your work accessible to a broader audience. Try it, and see how AI can simplify your translation and captioning needs!

Nikon D5 headphone and microphone jack

I am breaking down the topics you need to cover to start shooting a video with a DSLR. Part 1 will protect the essential gear you need. My assumption is you are interested in Storytelling.

The interview of a subject on camera is the staple of most storytelling. The gear I recommend will help you get a good discussion on video.

You can do many levels of production when recording video with your DSLR. I am taking the approach here to start you as a one-person band doing the show all yourself.

The starting point for all videos is sound. People will not watch a video with poor sound. Sound is the foundation for your production.

I recommend not trying to video unless you have a microphone and headphone jack. The microphones built into the camera would only really work for recording interviews within the selfie distance to the camera.

The basic understanding of recording sound for the video is the get the microphone as close to the sound as possible. I use a wireless lavalier system. The system I use costs $500.

Shure FP15/83 Combo Wireless System, H5

You can buy a wired lavalier for under $50. Buy one with a long cable for interviews.

The second principle to understand is that you need to isolate what is being recorded; the best way to do this is with closed-back headphones. Here is what I use.

Sennheiser HD 280 Pro Circumaural Closed-Back Monitor Headphones

You also need a good tripod to keep the camera still. A Pan & Tilt Fluid Head on the tripod is better than a bare head.

Google your camera using a microphone and headphones. Many cameras may have a microphone jack but not a headphone jack. There are some ways to work around that for your camera. Sometimes you will find you need to upgrade your camera.

You can add more gear, but I believe this is the basic kit necessary to create a video.

In other blog posts, I will discuss different microphones, lighting, and gear to help you move your camera during a shoot.

I will also cover the software you need to edit.

Stay tuned.

Last night I photographed, and video captured the show Oklahoma! Just a last-minute push to encourage you, if you are in Roswell, GA, to come and see the play. Runs through the weekend. Go here to buy your tickets or at the door.

Why I do, this is a great question. Most all DSLRs that record video have a 30-minute time limit. I understand this has to do with avoiding a double tax in some countries.

So how do you record a musical as I did that goes an hour and a half for just the first Act? This is where the Atomos Ninja Blade comes to the rescue.

Atomos Ninja Blade 5″ PRODUCT HIGHLIGHTS

Key Features

• 325DPI, 5″ IPS 1280 x 720 capacitive touchscreen monitor/recorder.
• Waveform RGB & luma parade, vectorscope with zoom, and test pattern generator.
• Adjustable gamma, contrast, and brightness.
• HDMI input and output.
• Real-time monitoring, playback, playout to a PC or Mac with QuickTime, and edit logging.
• Focus peaking, 0-100% zebra, and two modes of false color monitoring.
• Records 10-bit, 4:2:2 in ProRes or DNxHD.
• S-Log / C-Log recording.
• Trigger REC/STOP from camera (Canon, Sony, ARRI, Panasonic, RED, JVC)
• Timecode from camera.
• 2.5″ HDD/SSD media storage.

It records up to 1080 30p/60i resolution via HDMI to an available HDD or SSD using either Apple’s ProRes or Avid’s DNxHD codecs. Recording at 10-bit with 4:2:2 color sampling, this unit provides a monitoring and recording solution in one compact battery-powered unit.

I bought the ADATA Technology 256GB Ultimate SU800 SATA III 2.5″ Internal SSD card.

This setup worked great for last night’s opening night performance.

Fuji X-E2, Fuji XF 55-200mm, ISO 1600, ƒ/4.7, 1/100

While the Nikon D5 will record 4K, I don’t need this most of the time, so the Atomos Ninja Blade 5″ was perfect.

If you want to record at 4K, you can get the Atomos Shogun Flame 7″ 4K HDMI/12-SDI Recording Monitor. They make other higher-end models as well.

If you are not a video shooter and are more of a stills shooter, you need to use more hands-on techniques for capturing video.

What I mean by this is that the exposure and sound vary through a production; you may need to adjust this as you are recording.

With the Nikon D5 attached to the Atomos Ninja Blade, using the HDMI port, siphons this off before it hits the H.264 encoder, and you are recording in ProRes format. This isn’t registering in the RAW video but is more like a TIFF file than a JPEG.

Now I cannot share Oklahoma! Video because of copyright. [I am recording it for the Shuler Awards in Georgia]

The cool thing is you can now buy the Atomos Ninja Blade 5″ for $395 without a hard drive which gives you the 5″ monitor. I would recommend purchasing an SSD hard drive of your choice in size.

There are three more performances for me to tweak my exposure and sound to capture the best quality possible from my setup. At the same time, each version traditionally gets better each time.

By the way, Aunt Eller is my daughter Chelle. This is her senior year and last production. She also taught the choreography to the cast.

I hope these tips will help you think of ways to use your DSLR to do more than take photos. I wanted to use the high-quality CMOS chip to get an excellent keepsake video of our daughter to cherish for the rest of her life.

Print

When you shoot for print, especially a magazine, you will compose some photos for the cover. As a result, you leave a little more space than you would have if you had not been trying to make room for the text that needs to be on the cover.

Most editorial photographers benefit financially if the photo runs bigger and for how many images they use in the printed piece. Getting a double-truck [two total pages–side by side] also will give you more money. Here you have to consider the gutter, where the two pages are divided.

Sometimes they use one photo that bleeds off the pages or sometimes a grouping like here above. However, most designers are aware of the gutter and minimize this effect on the image.

Knowing you are shooting a photo story, most pros will do all they can to make the images so good that the editor feels the account must be on the cover and get extra pages. More variety helps the storytelling, and the photographer benefits from a more extensive paycheck.

The designers used to joke with my mentor, Don Rutledge; they thought he had templates inside the viewfinder to help him help them with the layout. But, really so many variables in situations that for any position, he usually had a cover shot, a double truck, some variation of the double truck, and then some tighter shots for impact.

Computer

Today the most likely place your photos will be on the web. The web is a horizontal space, and the general rule of photography is to fill the camera’s frame and watch the corners. The flat area also applies to your photo when filling the hole the designers are working with on the computer screen.

Here is how a vertical shot will look in the space for a designer. The horizontal layout for the desktop is the space it would fit within a video. The video format is about 16:9 in proportion.

Now the photo of the man here is the full 3:2 frame from the camera placed inside the 16:9 format with no cropping. While this is much better, the designer would prefer to fill their space, and so does the video editor.

I suggest shooting a little looser, giving some room for cropping. Some photographers will even put guides over the LCD screens to help them with their framing.

If you want to grow your business, be the guy that not only shoots for the use that the project is intended for but to meets other uses. Thinking beyond the immediate need is excellent stewardship because you can show the client that they can now use your photos for many different things, from brochures, newsletters, displays, PowerPoint, websites, and more.

Besides understanding the proportions, you need to consider what the audience uses to view your work. Knowing your audience is where your demographics can help you plan the coverage better.

You need to consider photos in a video on a website. Now, if your audience is mainly going to the web through a desktop or even laptop, you would shoot this differently than if they are using a Smartphone.

Shoot tighter if the experience is on a mobile device like an iPad or iPhone. For example, a panoramic shot with a person in the corner doesn’t read well on a 3″ screen. A smaller screen is where intimate photos of faces and objects are better for the space.

Think of how you tell time. Be sure the face of the clock fits the space to be able to tell the time.

As you can see, the photo of the watch makes it easy to tell the time.

As you can see in this photo, the watch and clock are about the same size. So if you want to see what time it is, get closer than this photo. In a video playing full speed on an iPhone, this image would lose its ability to communicate.

Are you asking the client, “What is the end use of the photos?” If not, you need to do this, and then you need to plan to be sure your photos make the best use of the space.