Re: Fast Five English Subtitles 720p Resolutionl
Vanina Mazzillo
LinkedIn and 3rd parties use essential and non-essential cookies to provide, secure, analyze and improve our Services, and to show you relevant ads (including professional and job ads) on and off LinkedIn. Learn more in our Cookie Policy.
All I wanted to make were videos of 30 seconds up to 1 minute of me talking to the camera or talking with someone else at the camera. I was publishing 'text' posts daily, so all I needed to do is to say 'that' on video.
I've documented the whole process step-by-step in this article to help anyone else who wants to put out more great looking video content on LinkedIn but feels held back by the 'traditional' process and the technology that normally goes into it.
I thought: what if I could make good videos much faster using just my mobile phone, so instead of posting 'text' daily, I could post 'videos' daily? How would that impact my personal brand and the company that I'm working for?
Definitely YES, and it's also very easy and fast once you grasp the basic techniques. You can shoot video on mobile with up to 4K resolution if your phone supports it, if not, anything above 1080 or 720 is enough. You can edit 4K videos on mobile as well. The audio recorded on mobile can sound great. You can add subtitles or captions and you can make thumbnails.
In short, you can do everything on mobile so your videos are great from a technical perspective. With that out of the way, you can then focus on the actual content that you will be producing, but that's for another article.
Next, I will go through each step of getting the video done in more detail, from start to finish. I will be answering the questions that I had in my mind while I went through this process. I will also explain exactly how I've done it and what tools I've used. Let's get some videos done!
In short, no. It's better to shoot videos with your native camera app and then upload that video on LinkedIn. By doing so you get better video quality and you also get the options to set up the format of the video, which you can't do in the LinkedIn app for now.
Another option is to get a dedicated app for shooting video, like Filmic Pro, but honestly, I did not see the point of doing that for LinkedIn videos. A dedicated video app makes sense for filming longer format videos and dynamic videos when you don't want the light to change while you're moving around. For a 30-seconds, static video, 'me talking at the camera scenario', I found that unnecessary.
This being said, LinkedIn videos that are not over-produced tend to have better performance in terms of views, likes, comments and shares. One way of getting that raw, personal vibe into a video is to hold it in your hand and get a bit of shakiness.
So far we've set up the front camera, we've put the phone down in a secure position and we are picking up great audio. Here are a few more aspects to consider before pressing the record button that can save you a lot of headaches and boost the quality of your videos further. Trust me, I've learned most of these the hard way!
A thumbnail is the first image that you see in the LinkedIn feed before pressing 'play' on a video. Typically this image has a text on it that explains what the video is about. Also, it's common to see the person in the video pointing their finger towards this text.
Think of the first frame of your LinkedIn video as being the Thumbnail. You'll need to edit your video and add text to that first frame. In this way, the users will clearly see what the video is about as they scroll through the LinkedIn feed, which exponentially increases the chances of them engaging with the video.
There are plenty of video-editing mobile apps that can be used for making a LinkedIn thumbnail. I tested KineMaster, InShot, and PowerDirector. I am personally using KineMaster but if my native camera app would let me add text on my video, I would definitely use that!
This 'bit of text' that you use to describe your video can literally make or break a video. If people don't press 'play' they'll never know what's beyond that point. Marketers and copywriters recognize the importance of a good title and they'll often spend more time on it than on making the article itself.
Similarly to Facebook where over 80% of users watch videos without audio, LinkedIn videos are also being consumed with no sound. So, by adding captions or subtitles your video gets a much better chance of being watched.
Finally, if you speak English with an accent as I do, then captions are a really good idea for your audience :) Remember, you want people to watch the video, and if they can't understand what you are saying, they will stop watching.
On the other hand, you CAN add a separate caption or subtitle file for your video when using LinkedIn on a desktop computer. When you're uploading the video on PC or Mac you get an option to add an SRT file, which in turn will add captions or subtitles to your LinkedIn video.
Since you can NOT add a separate captions or subtitles file in the LinkedIn app on your phone, the only option left is to hardcode them into the video. Hardcoding means using another app to make the captions or subtitles part of the video itself, effectively 'baking' them into the video.
I was wondering if it's possible to add some text (like movie subtitles) on the flow from VGA or HDMI. I may imagine that we would detect the resolution (i.e 1080p or 720p) and then add text on a specific pixel range.
I think you'd need a video interface or "video controller" chip. Maybe two chips... The one that goes on a video board and the one that goes in a monitor. Then, you might be able to control these chips and add pixels with the Arduino.
It's my understanding that HDMI from a Blu-Ray is encrypted. I'm not sure about other HDMI data. There's no way you can do anything with that (unless you can crack the encryption). i.e. If you are building an HDTV or monitor, you have to get a license and key.
I've done something similar to overlay telemetry onto a stream from a set of bullet cams. You can do it fairly easily with a PC that had a video capture card/dongle, an application to pull frames off the framebuffer and apply your captions and then write them to your output media stream, and a display adapter that supported your preferred video format/resolution. The ffmpeg library makes it quite easy to read/write video streams and manipulate them on the way through.
I wasn't aware that HDMI could be encrypted, I only thought It was like a "vga+sound" cable, no more, thanks for the tips! But for the chips I didn't really understand how to make it. Is there somewhere I could learn which kind of chips (video or other) I would have to use? Or just could you please guide me a little bit more so I could walk ahead on this subject?
So you need something with a video capture card, a substantial amount of processor power and RAM to manipulate frames in real time as they come off the card, a display adapter to convert the outgoing frames into your preferred video transmission standard, and presumably some sort of user interface to enable the captions to be defined or configured. Unless you plan to enter these on the fly while the video stream is being processed, that also implies persistent storage. Oh, and you need a fairly fast bus to tie this lot together.
Put that lot together and what you have is essentially what a PC is. Now, you can either spend a shed load of money buying the discrete components and then reinvent about 50 years of consumer electronic design to figure out how to make them all work together, or you can get an embedded PC (or even a standard domestic PC) that has all this ready and working for you. Which do you think is the smarter approach?
I clearly understand what you said and thank you for that. But, I'm not here to solve a problem with the smarter / easier solution. I'm here to understand how arduino is working and if this kind of idea could be handled by arduino.
It's certain that a PC could make it, ok. I just want to make it without a PC and to build one standalone system with the only hardware I need to. Why? Because it looks like fun to me and I'll learn tons of things working on this project. I'm not aiming to solve one issue as fast as possible. I'm here to learn how to build something specifically made for one idea.
Maybe arduino is just not made for this kind of project. Maybe there is something like arduino oriented to video. Or maybe it could be done with a lot of work and then show that arduino could make everything.
I don't see why you have to capture the video, all you have to do is to generate your captions synchronised to the incoming video and then over lay them using a mixer or video switch.
It is still quite a project and not one for a beginner. The arduino part is the least of it.
The Arduino compatible shield board overlays text on analog video, using the Maxim MAX7456 on-screen display chip. This board allows easy overlay of text onto an analog video signal, NTSC or PAL. Connect this board to your favorite Arduino processor...
I've never seen one of those before. That looks like the perfect solution to jinsa's problem. They aren't cheap, and seem to be out of stock at the moment, but if you can get your hands on one it should be just what you need.
You could try modifying a NeTV. It overlays information over HDMI video without breaking DMCA (should be checked with legal though). If you want to overlay, you just need to learn to write applications for NeTV: it's an overlay solution for HDMI (even HDCP encrypted).
I stored each lines info ( or several lines repeated to make the blocks ) during the horizontal flyback and first part of the trace, into 2 parallel in / serial out registers, which I read out to make the video signals to add to the video.
ffdshow is a decoder-cum-post-processing package that is common in codec packs such as CCCP and K-Lite Codec Pack. It not only can decode various video and audio formats, but also do some post-processing on them. Post-processing allows you to improve picture quality, repairing bad ones and some other features like logo scrubbing etc.
7fc3f7cf58