My new favorite media player creates and translates subtitles on the fly, and I'm saying goodbye to SRT files

Published Jul 19, 2026, 10:30 AM EDT Gaming has been Samarveer’s greatest passion, and the Literature graduate in him takes immense joy in dissecting games for their themes, messages, and impact. Samarveer holds a deep appreciation of gaming, and considers the platform to be the most immersive and impactful across all media. He can be found engaging with gaming communities online, always ready to debate the finer points of ray tracing or itching to write an 8-page collegiate thesis on any game that impacts him emotionally. Summary I stopped hunting for SRTs - PotPlayer builds real-time subtitles from audio using Whisper. It translates foreign dialog live (Whisper + Google Translate), making foreign media instantly watchable. Easy setup: right-click Subtitles -> Create subtitles from audio, pick Vulkan and a Whisper model. For me, downloading subtitles for a new media file has always been "part of the ritual." If the timings match after pulling a subtitle file from a website, great, but if they don't, then we try again. This part is never difficult, but it sure can get annoying enough to interrupt the experience. Even today, plenty of media players still expect you to do that work yourself, and sure, a lot of us do it, too. However, ever since I stumbled upon PotPlayer and finally replaced VLC, the subtitle-hunting process now feels outdated to me. Instead of searching for any subtitle files at all, I simply let PotPlayer generate them on the go, and this feature helps me out in more cases than just one. PotPlayer creates subtitles while you watch It feels like magic until you realize what's happening For the past couple of months, I've genuinely been impressed by PotPlayer's laundry list of features that it boasts over the rest of the media players in the business. However, I didn't expect real-time subtitles to be one of them, and for it to work this well, too. Naturally, I'd assumed that PotPlayer was downloading subtitle files from somewhere in the background, like the Bazarr integration on my Jellyfin server. However, it actually listens to the video's audio track, converts speech into text, and builds an entirely new subtitle track on my PC as the movie plays. This is far from just a convenience feature, and comes off as a genuinely smarter way to handle subtitles. The magic is accomplished thanks to OpenAI's Whisper speech recognition model. PotPlayer can tap into Whisper to transcribe spoken dialog in real time, after which it generates subtitles without the need for a pre-existing SRT file. Since everything is based on the audio itself, it doesn't matter whether the video is a decade-old download, a random clip from my archives, or something that never had subtitles to begin with. Now, to be completely fair, the result isn't completely flawless, but it's remarkably accurate on clean, well-mixed audio. PotPlayer's real-time subtitles do not require you to manually download any AI models or engines. All downloads are handled by the media player itself. To use this feature, simply load up any media file (preferably with English audio) into PotPlayer, and right-click anywhere on the window. Hover over Subtitles to bring up the subtitle sub-menu, and click on Create subtitles from audio (Real-time). From here, select the Vulkan option in the Whisper AI Engine menu, and then choose a Whisper AI model from the provided list. Pick the "tiny" models if you're using PotPlayer on lower-end hardware, and larger, billion-parameter Whisper AI models if you want the highest accuracy and are willing to spare 5–6 GB of memory to PotPlayer. Lastly, let Whisper AI language remain set to Auto. Potplayer will take a couple of minutes to download the model you've picked (more if you go with the larger models), and once it's done, you will have real-time subtitles in your videos. It also breaks language barriers like they're nothing Real-time translation makes foreign media far more accessible Generating subtitles from real-time English audio is impressive enough, but PotPlayer's translation feature is the one that genuinely sold me. Instead of merely transcribing what's being said, it can listen to dialog in another language, recognize the speech, and generate English subtitles on the fly. Thanks to that, I no longer find myself searching for regional subtitle files before watching foreign content because my media player itself handles that part for me. Sure, it does take some context clues to make a quick sanity check about everything lining up, but just a few scenes later, things begin flowing along smoothly. Once again, Whisper does most of the heavy lifting here, with help from Google Translate. You can even select other translation engines like Bing, Libre, and LM Studio. It recognizes spoken dialog, identifies the source language, and converts it into English subtitles as the video keeps playing. It's genuinely impressive how fast this process happens, too, because after a few seconds, it just begins to feel like the media file came with subtitles embedded from the get-go. The quality of PotPlayer's real-time caption depends heavily on the audio quality of the video. Of course, there are drawbacks involved, like there usually are with any speech recognition system. The results depend heavily on the audio quality. If I'm watching professionally sound-mixed films, documentaries, and TV shows, I will most probably get excellent real-time translations, thanks to the dialog being clean and isolated. On the other hand, my home videos with crowded environments or inconsistent microphone quality still trip up PotPlayer's inbuilt translator. For all sorts of polished media, though, I've found it nothing but dependable. For this setting, right-click anywhere on the video player's window and click on Subtitles again. This time, click on Subtitle Translation, and select Google Translate as your default translation engine. I've found selecting the Show translation only option helps keep things tidy on the screen, so toggle that option while you're there. Near the bottom, you can also find options to manually select the Source Language and Translation Language to help the AI engine work better. Seconds later, your real-time regional audio, too, should be showing up with mostly-accurate English subtitles. It removes a huge, frictional pain-point from the media-watching process It's remarkable how I can't remember the last time I manually downloaded subtitles after downloading PotPlayer. I've genuinely saved time on movie nights and managed to keep the vibes intact by letting PotPlayer handle subtitle delivery instead of asking guests to wait while I search for the right SRT file. All I have to do on my end now is press Play and assume everything will work itself out, which, nine times out of ten, it does, thanks to PotPlayer's incredible capabilities. That's ultimately why PotPlayer has earned a permanent place on my PC. The AI features don't try to steal the spotlight, nor are they marketed loudly from the rooftops. They just exist to help me solve a decades-old problem, and once they do that, they get out of the way. PotPlayer understands what me and my family are watching, and it adapts accordingly, making it genuinely impossible for me to ever consider going back to VLC ever again now. PotPlayer PotPlayer is the VLC alternative you didn't know you needed, with its litany of options that suits a power user perfectly. This is the kind of AI I actually want Why isn't every bit of AI tech used exactly this way? Whisper AI's models are integrated directly into PotPlayer, and all it does is call the right model to take over locally. These real-time subtitles and translations, too, make language barriers and subtitle hunting feel like problems from a bygone era. I still appreciate having manually curated subtitle files with a random "(&nbsp)" thrown in randomly for good measure. However, ever since I began using PotPlayer was instant and accurate real-time subtitles, I feel liberated. I no longer have to depend on the right subtitle file or website, which is a level of convenience I had never thought I'd reach.