Subanana
Accurate subtitles for the languages other tools get wrong
Start new thread

Live Captions by Subanana - Your whole audience follows, in their own language

by
•
You speak once. Your audience scans a QR code, picks a language, and follows you live on their own phone — reading it, or hearing it. Put two languages on the big screen, or send a clean feed into OBS and vMix.

Add a comment

Replies

Best

Good launch, congrats!

We have a constant problem with Armenian-language transcription =)

I’ll definitely give it a try!

Maker

 Thank you! And good news — Armenian is well supported, so you're in the right place. Do give it a try on a real recording. If anything about the Armenian output could be better, tell me directly — that kind of feedback is exactly what we act on.

I’ve already signed up! =)

And does it support links from YouTube or other platforms?

Maker

 Welcome aboard! And yes — paste a public YouTube link and we'll fetch and transcribe it, no downloading first. Instagram and Facebook links work the same way. Or upload the file straight from your machine if you'd rather — video or audio, up to 30 GB, on every plan. Both feed into our meeting transcript and subtitling tools, so you get the transcript and the subtitles from the same import.

 I have some feedback, and I'd like to share it via LInkedin or somewhere =)
Subanana is really good (Forgot to mention that i liked the idea of it's name Subanana). Congrats again!

Maker

 Thanks!

Maker

Hey Product Hunt — Aric here, from the Subanana team in Hong Kong.

We build in a city where a room is rarely all one language. The usual fix at an event is an interpreter booth, a stack of receivers, and a budget most organisers do not have. So we built the version that runs on the phones people already brought.

Here is what it does. You start a session and speak. Your audience scans a QR code, picks their language, and follows along on their own screen — reading the captions, or hearing the translation spoken. You can run up to five languages at once, put two of them on the big screen, float a subtitle box over your slides, or drop the captions into OBS, vMix or ProPresenter as a layer with a see-through background.

Two things worth knowing before you try it. It listens to your microphone or your computer's audio — it does not join a Zoom or Teams call as a bot. And the captions appear a moment late on purpose: we clean up the text before you see it.

Every plan including the free one gets a 5-minute live session, so you can point it at a real meeting today. Live captions themselves are on our Max plan.

If you want to go further, the code PRODUCTHUNT20 takes 20% off your first three months. It works until 10 October.

If you run events where the audience does not share one language, I would like to know where your current setup costs you the most.

 How well does it handle mixed-language conversations in the same sentence ?

Maker

 That's the case we built this for. Hong Kong (where we base) meetings switch between Cantonese and English mid-sentence all the time, so it had to work. Live captioning listens across your languages at once and keeps the switch as it was spoken, instead of forcing the whole sentence into one language. Best test is a real conversation rather than a demo — give it a go and tell me how it holds up.

Maker

Hey everyone — Kai here, one of the builders on Subanana. Adding a bit of the "why" behind a few decisions Aric mentioned:

Why we route per language. Our first product taught us that no single speech model is best at everything. Great English accuracy often turns into mangled Cantonese or messy code-switching. So Live Captions uses the same routing as our subtitle tool: each language goes to the model that handles it best, not whichever vendor we happen to be locked into.

Why the captions are a moment late. This one was a debate for us. Raw real-time output flickers, rewrites itself, and breaks mid-sentence, which is tiring to read on a phone. We chose a short delay so we can clean up the text before anyone sees it. A slightly later caption you can trust beats an instant one you have to re-read.

Why no meeting bot. It listens to your mic or your computer's audio, so it works the same on a conference stage, in a church, a classroom, or next to a Zoom call. You don't need to invite anything into the meeting.

Built for the production desk too. Beyond the QR code on phones, you can run up to 5 languages at once, put 2 on the big screen, float a subtitle box over your slides, or send a transparent caption layer into OBS, vMix, or ProPresenter.

Every plan, including free, gets a 5-minute live session, so you can try it on a real meeting today.

I'm hanging around all day. Happy to answer technical questions, and I'd love to hear which language pairs you'd want us to stress-test next. 🙏

Maker

Hey everyone! I am so glad to be a part of this launch! I am Lakshya, a friend of Aric and Kevin. When Aric told me about this new feature of Subanana, I wasn't sure what it really did. But when I understood, I was surprised no one had built this before. Such an obvious pain point in hindsight!

I have been to multiple events where the speaker was talking in a different language and I felt lost.

Using Live captions by Subanana, any event can now become something everyone can follow and enjoy.

I am helping them market this feature better because I know it deserves more eyeballs.

Huge congrats to the whole Subanana team!

Maker

Co-founder here. Thanks for checking us out, and huge thanks to Aric and Kai for getting this out the door.

The thing I care most about: the person in row 30 who doesn't speak the stage language. No headset, no booth, no app to install. Scan the QR, pick your language, follow along on your own phone.

Every plan, free included, lets you run a 5-minute live session. Please try it with a real mic and a real language pair and tell us where it breaks. That's the feedback we'll act on first.

Never thought I needed something like this but Live captions now feel such a necessary addition in any event now. Congrats Subanana team on your launch.
Maker

 Thank you! That's the reaction we hoped for. Most people don't think about captions until they're in a room where part of the audience is quietly lost. Every plan includes a 5-minute live session, so you can point it at your next event and see for yourself. Would love to hear how it goes.

This is exactly the product that I've been dreaming for. This is especially going to be useful for Georgian language, as transcription for Georgian content is usually problematic.

Maker

 Thank you — and good news, Georgian is in the language list, so you can pick it and go. Under-served languages are the whole reason we built this, so please run a real recording through it and tell us how it does. If the Georgian output could be better anywhere, say so here and we'll act on it.

Maker

Wow Aric! It's truly awesome cause online events usually gets viewers from different countries and it makes them more accesible and friendly. Wish you all the best here

Maker

 Thank you Germán. That's exactly it — the audience is usually more mixed than the host realises, and most of them just quietly follow along at 70%. Making the room easier to be in is the whole point. Appreciate the kind words.

Congrats on the launch, been on the beta and it clicked at a demo meeting last week:

Hong Kong client, local team + two people over from the Shenzhen office, and the client went "don't mind them, carry on in Cantonese", which is always awkward.
Put the QR up, they read it in Simplified on their phones, done.
I kept the Cantonese line up next to the translation and it actually follows how people talk here, half Cantonese half English, instead of rewriting it into written Chinese or mangling the English words like every other tool I tried.

Having worked on AI products myself I know how little Cantonese data there is to build on, so genuinely impressed you pulled this off.

Maker

 Thank you Nicholas, and thanks for being on the beta. That meeting is exactly the room we built this for. The awkward moment isn't the language, it's everyone waiting while someone half-interprets — and a QR code on the screen removes it without anyone changing how they speak. The mixed Cantonese-English line was the hard part and the part we refused to compromise on. It would have been much easier to clean it into written Chinese. That's not what was said in the room, so we don't do it. You're right about the data. Getting there took a lot of stubbornness. Coming from someone who has shipped AI products, this means a great deal.

Maker

Hey! Engineer of the Subanana team here 👋, and the situation I keep picturing is the handoff halfway through a multilingual event: the keynote ends, a guest speaker takes the stage in another language, and part of the audience suddenly needs a different way to follow along; it is the kind of ordinary transition that looks simple on a run sheet but can quietly leave part of the room behind.

One detail we built for that reality is the ability to change the source and audience languages through Edit event during an ongoing session; saving briefly pauses and reconnects the captions, subsequent speech uses the new settings, and the earlier caption history remains unchanged, which matters at panels, lectures, community gatherings, church services, and mixed-language Q&As where the language mix is rarely as tidy as the agenda suggests.

covered a lot of the thinking behind the live experience, so I’d love to hear from people who run these rooms: when your event shifts from one speaker or language to another, what creates the most friction—the handoff itself, getting the right language to each part of the audience, or keeping specialised names and terms consistent throughout the session?

12
Next