How to Add Translated Subtitles to Zoom Calls
TL;DR: StreamVox overlays live, translated subtitles on top of a Zoom call, or a call in Teams, Meet, Discord or any other app that plays audio. It works on any Zoom plan, in any of 69 languages: you read the call in your language, and when the people on the other side install StreamVox too, each of them reads it in theirs. Set your input and output language, turn on the overlay, and subtitles appear as people talk.
StreamVox is a real-time AI subtitle app for Windows, Mac and iPhone. On a computer it captures whatever audio is already playing, system-wide or from a single app, and shows a floating, translated overlay on top. That means a Zoom call gets live translated subtitles regardless of which Zoom plan is on the account, because the translation happens outside Zoom entirely.
Does Zoom Have Built-In Translation?
Yes. According to Zoom's official help (checked in October 2026), Zoom has translated captions: speech in a meeting is turned into captions in another language in real time. They require the host to be on a Zoom Workplace Business Plus, Enterprise Essentials, Enterprise Plus or Enterprise Premier account, or to have the Zoom Translated Captions add-on. The host can choose which language pairs are offered, and each participant turns captions on and picks a language. Zoom also has a Voice translator, which speaks the translation aloud and needs the Live Translation add-on or ZoomMate.
StreamVox works another way, and you can use it alongside Zoom's features or on its own. It runs on your own Windows PC or Mac, listens to the call audio you already hear and shows translated subtitles in an overlay on top of Zoom. Nothing has to be turned on in Zoom, by you or by the host, and any Zoom plan works. It recognizes speech in 61 languages and translates into 69, and the same overlay works on Teams, Meet and Discord. On iPhone it cannot subtitle calls, Zoom calls included: iOS does not pass call audio to third-party apps such as StreamVox.
How to Set Up Real-Time Zoom Translation (5 Steps)
Getting subtitles running on a call takes about a minute:
- Download StreamVox for Windows or Mac and sign in.
- Join your Zoom call as usual, in the desktop app or a browser.
- Pick your languages - the language you'll hear, and the language you want to read.
- Choose your audio source. In the Sound: menu, System Audio works for most people; pick Application and choose Zoom to isolate the call if you have other audio playing. On Windows, if Zoom's audio isn't being picked up, switch to System Audio and turn on If a call app is silent at the bottom of the same menu: it's built for exactly this. See the full troubleshooting steps if it still doesn't register. On a Mac there is no such option: StreamVox captures meeting audio directly.
- Turn on the overlay and start talking. Translated subtitles appear as the meeting audio comes in.
That's the whole setup. The Zoom and Teams translation use case page walks through the overlay in more detail if you want to see it before installing.
Does It Work with Teams, Meet, and Discord?
Yes - the overlay doesn't care which app is making the sound:
- Zoom, desktop app or browser
- Microsoft Teams
- Google Meet
- Discord voice calls
- Webex
Because StreamVox captures audio at the system level rather than plugging into any one app, switching from a Zoom call to a Teams call doesn't require a different setup - the same overlay, languages, and settings carry over.
Can Both Sides Talk in Their Own Language?
Yes - and the solution is simpler than most people expect: every participant installs StreamVox on their own computer. Each copy works independently - it translates the incoming call audio into whatever language its owner picked. Your colleague speaks English and you read Spanish; you answer in Spanish and their copy shows it to them in English; a third teammate joins with a German output and reads the same meeting in German. Two, three, or ten participants - there's no limit, because nobody shares anything: each machine just translates what it hears. Nobody has to switch languages or slow down for the other side.
One thing StreamVox's overlay can't do is appear on someone else's screen, so if the other side needs to read you, the answer is their own copy of StreamVox, not a setting on yours. Conversation mode, despite the name, solves a different problem: following both languages of one conversation on your own screen. Point it at the microphone, your device's audio, or, on Windows and Mac, a chosen app, and it takes any two of 10 curated languages, detecting the spoken language automatically. That covers an in-person interview or a reception-desk chat just as well as a bilingual voice chat in an online game. On a call, Windows and Mac can take the microphone and the computer's sound together instead (wear a headset): you speak into the microphone, the other side arrives through the computer, and each side is recognized on its own channel. The 10-language limit then disappears entirely: you get all 61 languages StreamVox can hear, and Quality mode works fully. On iPhone that second setup is not available, so Conversation there is always the curated 10. The complete guide to two-way AI translation covers both setups in depth.
For the meeting itself, the Meeting translation profile is tuned for professional and industry terminology, and Quality mode is there when you want the most accurate read of a call over raw speed. It's the same setup that makes a live-subtitled international interview work when the call is a job interview or a client negotiation instead of a team standup.
Is It Private?
Audio is processed in real time to generate your subtitles and is never stored - not by StreamVox, and not by the AI providers it uses. That applies whether the call is a client pitch, a performance review, or a casual team standup. StreamVox doesn't join the meeting as a bot and doesn't install anything into Zoom - it just reads the audio already on your own machine. That's worth keeping in mind: if you're translating the other side of a call, do it the way you'd want it done to you, and let colleagues and clients know you're using it when it's relevant to the conversation.
What Does It Cost?
The free plan gives you 15 minutes of translation a day - enough to try StreamVox on part of a call, not to cover a full hour-long meeting start to finish.
Pro is $8.99/month for 40 hours, which works out to roughly two hours of meetings on every workday. Pro+ is $14.99/month for 70 hours, and Unlimited is $24.99/month with no cap. Every plan includes the same overlay, translation modes, Conversation mode, and 61 languages - the only difference is how many hours you get.
Frequently Asked Questions
Does this replace Zoom's built-in translated captions?
No. It works another way, and you can use it alongside them or on its own. According to Zoom's help article on viewing captions in another language (checked in October 2026), translated captions need the host to be on a qualifying Zoom Workplace plan or to have the Zoom Translated Captions add-on. StreamVox runs on your own computer, needs nothing turned on in Zoom and works with any Zoom plan. It translates into 69 languages and works the same way on Teams, Meet and Discord.
Why isn't Zoom or Teams audio being picked up on Windows?
Some Windows communication apps route audio through a path StreamVox's default capture mode can miss. Choose System Audio in the Sound: menu, turn on 'If a call app is silent' and restart the capture session. The full troubleshooting steps are in our help center.
Can both sides of the call talk in their own language?
Yes: each participant runs StreamVox on their own computer. Your copy translates what you hear into your language; your colleague's copy translates your voice into theirs. Everyone picks their own language pair, and it scales to any number of participants. (Conversation mode solves a different problem: following both languages of one conversation on your own screen. It takes 10 curated languages from a single audio source, which is always the case on iPhone, or all 61 languages it can hear, with Quality mode working fully, on Windows and Mac when you combine the microphone with computer sound on a call.)
Is meeting audio recorded or stored?
No. Audio is processed in real time to generate subtitles and is never stored, not by StreamVox and not by the AI providers it uses. Nothing about the meeting is saved after the subtitles appear.
How much of a meeting can I translate for free?
The free plan gives 15 minutes a day - enough to try StreamVox on part of a call, not a full hour-long meeting. Pro, at $8.99/month, unlocks 40 hours a month, about two hours of meetings on every workday.
Get Started
- Download StreamVox for Windows or Mac.
- Sign in, pick your input and output languages, and set your audio source to System Audio (on Windows, also turn on 'If a call app is silent' if Zoom stays silent).
- Join your call and turn on the overlay - translated subtitles appear as people talk.
- If the other side needs to read your language too, they install StreamVox on their computer and pick their own language pair - it scales to any number of participants.
- Upgrade if you need more than 15 minutes a day - Pro, Pro+, and Unlimited unlock the same overlay and modes with more hours.
Whatever Zoom plan the meeting runs on, you can follow it in your own language. Point StreamVox at the call and read along from the first word.
AlekGir
Ready to Try StreamVox?
Get real-time translated subtitles for any audio on Windows, Mac and iPhone.
