AI transcription tuned for Cantonese
Speech is transcribed into a clean, punctuated Cantonese transcript first, because everything the translation gets right depends on getting this step right.
Translate Cantonese video to English text online free. UniScribe transcribes your Cantonese video, then translates it into English text you can export.

The Cantonese source and English target are already selected, so just drop in the Cantonese video you want to translate to English text.

UniScribe transcribes the spoken Cantonese into an accurate transcript, then translates that transcript into English text.

Edit the English text, then export it as SRT or VTT subtitles, or as TXT, DOCX, PDF, or CSV.
Cantonese is written without spaces, marks time with particles instead of tenses, and its spoken form differs from the written Chinese you may have learned. English needs all of that made explicit. That is why UniScribe splits translating Cantonese video to English into two steps: it transcribes your Cantonese video into an accurate transcript first, then translates that transcript into English text you can edit, export, and share. Translate Cantonese video to English text in minutes, free to start.
Source language
Target language
Cantonese-to-English translation has to make explicit what the particles and the context carry. Four differences shape every result.
Cantonese is written without spaces, so the transcript has to decide where each word begins and ends before anything can be translated.
Verbs never change form. Small particles after the verb say whether something is done, ongoing, or experienced.
Particles at the end of a sentence add attitude, certainty, or doubt, and English has to express that with wording instead.
Everyday Cantonese uses its own words and characters that differ from standard written Chinese, so the transcript has to capture what was actually said.
See it in action
Cantonese
呢份報告我哋上個禮拜傾咗喇,下個月再睇吓效果。
English
We already discussed this report last week; we'll see how it performs next month.
This example highlights:Particles mark time, not tenses · Final particles carry the tone · Spoken Cantonese isn't written Chinese
Those differences are why a single word-for-word pass falls short. UniScribe is built to translate video accurately: it translates Cantonese video to English in two AI steps, transcribing the Cantonese video first, then translating that full transcript into English with the surrounding context intact.

Speech is transcribed into a clean, punctuated Cantonese transcript first, because everything the translation gets right depends on getting this step right.
The whole transcript is translated with its context, so English tense, subjects, and word order come out as real sentences rather than a literal gloss.
Transcribe in 60+ languages and translate between them, so the same workflow covers whatever recording lands on your desk next.
Meetings and interviews stay readable. Each speaker is labelled, so the English transcript still shows who said what.
Download the English text as TXT, DOCX, PDF, SRT, VTT, or CSV, whether you need a document, subtitles, or a shareable link.
Turn a long recording into key points, action items, and mind maps so you can act on it without reading every line.
A video translation rarely ends as plain text: it usually has to go back on top of the footage. UniScribe keeps the timing, so the English text is ready to use as subtitles.
Export the English translation as SRT or VTT with the timecodes intact, ready to load into any player or editor.
Translate YouTube and other video links directly, with no download or conversion step in between.
Upload MP4, MOV, WEBM, and more; the audio track is extracted and transcribed automatically.
Every segment keeps its original timing, so you can jump from the English text back to the exact moment in the footage.
Cantonese meetings and calls become English minutes with the decisions and follow-ups written down.
Cantonese footage becomes English text you can build subtitles and scripts on top of, starting from what was actually said.
Focus groups and interviews recorded in Cantonese, read by an English-speaking team without waiting on an agency.
Cantonese recordings preserved as English text that the whole family can read and search.
From short voice notes to long-form recordings, people around the world use UniScribe to turn audio and video into text they can actually work with.
UniScribe has become a part of my daily workflow, and I believe it should be in every content creators' toolbox. It's the best transcription tool I've ever used. The accuracy is amazing. The summary and FAQs that it creates with every transcription was a nice surprise, they allow me to actually create a valuable add-on to every podcast and video summary I upload to my membership site in not time at all. This has been one of my favorite AppSumo purchases. I give it 5 tacos! If you're on the fence about whether or not to purchase a UniScribe subscription, get off the fence and do it. It's a great value.
I’ve been searching for an alternative to subscription-based transcription apps and recorders, unsure if they’d truly boost my productivity as a teacher. UniScribe does exactly what it promises! I’ve recorded various things using the Voice Memo app on my phone, and simply dropping the file into UniScribe gives me a seamless transcription. I can file it, export it, and move on quickly. I’ve even created a shortcut on my iPhone for direct Chrome login, making access even faster. I had a minor login hiccup, but their support team resolved it quickly. The YouTube transcription feature is next-level – perfect for quick content checks as a teacher.
This tool is a key part of my stack. And the recent update where we can generate different summary types from the same transcript has taken the tool to another level! My key reason when I picked up this tool was to make mind maps from the videos. And this tool keeps delivering on that. The processing speed, quality of transcript, multiple sources, and ability to replay audio later—I mean, can it get more valuable than this?
Effortlessly transcribe audio and video, saving you time and helping you focus on what matters