Dubbing is not switched on yet. You can look around and create your account now.

24 languages · every sentence checked

AI dubbing that shows you what to check

DUBBE dubs your video (a vlog, a lecture, a film) into another of 24 languages (Korean ↔ English tested most), keeps the background sound, and marks the doubtful sentences, so a person can fix them before anyone watches.

Start freeSee an example

Made for creators · films & series · lectures · e-books, read aloud by Text to Speech

Built on open models

Whisper large-v3MADLAD-400Qwen3Ming-3.0CosyVoice 3Kokoro-82MDemucsChatterbox MultilingualNemotronTitaNetemotion2vecAudioSealfaster-whisperCTranslate2FFmpeg
The real example, playing

Every sentence, checked as it plays

Korean read aloud, dubbed into English with no human correction. Each line shows the original, the dub, and the flag DUBBE put on it when it was unsure.

15
sentences dubbed
5
flagged to check
2.7
minutes of video

Open the full example

01 · Hear

Every word heard, twice

Whisper writes the speech, a second recogniser listens again, and sentences the two hear differently are marked for you. Speakers are told apart by their voices.

Whisper large-v3Qwen3-ASRSpeakers by voice
02 · Translate

Meaning carried across

MADLAD translates each sentence; a judge model scores it, and numbers and your glossary terms are checked. Doubtful sentences get a flag on the timeline.

MADLAD-400MetricX-24Glossary
03 · Speak

A new voice, on time

Each speaker gets a voice; a sentence that comes out wrong is heard back and said again; and each one starts where the original was said — with the music and sounds of the original kept underneath.

Ming-3.0CosyVoice 3Heard back

Start free

AlloyAmerican · 138 Hz
AoedeAmerican · 157 Hz
BellaAmerican · 123 Hz
HeartAmerican · 159 Hz
JessicaAmerican · 175 Hz
KoreAmerican · 123 Hz
NovaAmerican · 156 Hz
RiverAmerican · 170 Hz
SarahAmerican · 162 Hz
SkyAmerican · 130 Hz
AdamAmerican · 116 Hz
EchoAmerican · 110 Hz
EricAmerican · 152 Hz
FenrirAmerican · 130 Hz
LiamAmerican · 115 Hz
MichaelAmerican · 106 Hz
OnyxAmerican · 90 Hz
PuckAmerican · 111 Hz
AliceBritish · 180 Hz
EmmaBritish · 173 Hz
IsabellaBritish · 171 Hz
LilyBritish · 140 Hz
DanielBritish · 114 Hz
FableBritish · 128 Hz
GeorgeBritish · 126 Hz
LewisBritish · 91 Hz
AlloyKorean, from the American preset · 137 Hz
AoedeKorean, from the American preset · 154 Hz
BellaKorean, from the American preset · 170 Hz
HeartKorean, from the American preset · 170 Hz
JessicaKorean, from the American preset · 177 Hz
KoreKorean, from the American preset · 134 Hz
NovaKorean, from the American preset · 147 Hz
RiverKorean, from the American preset · 160 Hz
SarahKorean, from the American preset · 163 Hz
SkyKorean, from the American preset · 143 Hz
AdamKorean, from the American preset · 111 Hz
EchoKorean, from the American preset · 104 Hz
EricKorean, from the American preset · 136 Hz
FenrirKorean, from the American preset · 118 Hz
LiamKorean, from the American preset · 115 Hz
MichaelKorean, from the American preset · 114 Hz
OnyxKorean, from the American preset · 91 Hz
PuckKorean, from the American preset · 105 Hz
AliceKorean, from the British preset · 167 Hz
EmmaKorean, from the British preset · 165 Hz
IsabellaKorean, from the British preset · 174 Hz
LilyKorean, from the British preset · 160 Hz
DanielKorean, from the British preset · 115 Hz
FableKorean, from the British preset · 112 Hz
GeorgeKorean, from the British preset · 124 Hz
LewisKorean, from the British preset · 99 Hz
24 languages, any pair

Pick a language, any language

Every language below can be the original or the dub. Korean ↔ English is tested in depth; the others are measured on read speech (FLEURS) — see the results.

  • Arabic
  • Chinese
  • Danish
  • Dutch
  • English
  • Finnish
  • French
  • German
  • Greek
  • Hindi
  • Italian
  • Japanese
  • Kazakh
  • Korean
  • Malay
  • Norwegian
  • Polish
  • Portuguese
  • Russian
  • Spanish
  • Swahili
  • Swedish
  • Turkish
  • Uzbek
Dub intoArabicfrom any of 24
Why flags

Stop listening to every sentence

Without flags

Listen to all 15 sentences to find the mistakes

Every sentence looks the same until you have heard it.

With DUBBE

Start with the 5 flagged sentences

5 of 15 on the example clip. Then one final listen — flags miss confident mistakes.

What you get

Everything a lecture dub needs

Review flags on the timeline

Translation, speech recognition and timing problems, coloured where they happen.

Background sound stays

Music and effects duck only while the new voice speaks.

 

Subtitles in both languages

SRT and WebVTT, timed to the dubbed video.

…forests in Ghana…

Fix a word, re-dub

Corrections are never re-translated; re-dubs use no minutes.

0.00 s

No drift

Every sentence starts where it was said — measured on three clips.

Delivery tags

Slower, softer, a question, a pause — per sentence. DUBBE also tags what it hears in the original (questions, emotion, laughs, sighs); you keep or untick them.

Sentences
15
To check
5 of 15
Kept as
a web page

A report for every video

Time, cost and every check, as a page you can keep.

How it works

Upload, wait, review

30 s to 1½ min of work per video minute. The thirteen steps of every dub, as they run:

dub · lecture.mp4 · ko → en 13 steps
  1. Hear
  2. 01Extract sound
  3. 02Separate voice
  4. 03Recognise speech
  5. 04Find sentences
  1. Understand
  2. 05Tell speakers apart
  3. 06Hear the delivery
  4. 07Pick voices
  5. 08Translate
  1. Speak
  2. 09Speak
  3. 10Shorten
  4. 11Time to picture
  5. 12Mix
  6. 13Report
01

Upload

MP4, MOV or MKV in one of 24 languages.

02

DUBBE dubs and checks

Thirteen steps: it tells the speakers apart, gives each a voice, and checks every sentence.

03

You fix and download

Video, subtitles in both languages, a transcript by speaker, a report.

New this week

Several speakers, each with their own voice

Speakers told apart

Every sentence is matched to a speaker by voice; each speaker gets a different voice, and the transcript names them. Similar voices can be taken for one person — split them in the editor.

Their own voice, with consent

A speaker can be cloned from the video only after you confirm they agreed. The copy stays in that project, or private in your voices — never published.

The delivery, heard

Questions, emotion, laughs and sighs in the original become tags on the sentence, shown dashed in the editor so you keep the right ones.

How it works →

New · Voice Marketplace

Speak with voices people chose to share

Clone your own voice in about a minute, publish it and see who uses it — or pick a voice someone else published for your speech and dubs. 0 voices listed so far.

  • Only the publisher's own voice, recorded live and checked by speech recognition
  • Every listing heard by a person before it goes live
  • Watermarked and labelled as AI wherever it is used

Sign up to browse

Made for

Videos where a wrong sentence matters

Cost of a course

How much video do you need dubbed?

A semester of 30 lectures × 75 min is 2,250 minutes.

DUBBE processing time
DUBBE GPU cost
ElevenLabs automatic dubbing
Human studio dubbing

Measured 29 s per video minute into English on an RTX 4060 (77 s into Korean); GPU at an assumed $0.35/hour. ElevenLabs ≈ $0.50–0.60/min from its credit prices; human dubbing $80–250/min is an unverified industry figure. Bars on a log scale. During the beta DUBBE is free (15 min/month).

Measured, not promised

9.1%

character error rate of speech recognition on 22 minutes of read Korean (FLEURS, one pass)

29 s

of processing per video minute into English on one laptop GPU (Korean: 77 s)

0.00 s

largest timing delay on three test clips (one with picture slow-down)

24

languages, in any pair — each with 26 voices, picked per speaker by pitch

All results — including what does not work well yet →

Honest limits

Narrow on purpose

No lip sync; people in your video are cloned only with their consent; Korean ↔ English is tested in depth, the other 22 languages only on read speech (FLEURS). DUBBE is built for lectures, where one wrong sentence matters.

Safety, built in
  • Upload only what you own or may change — confirmed on every upload
  • Your own voice after a live check; a speaker in your video only with their consent, and never published
  • Every dub is labelled AI-generated, in the file and on the download page

How we handle safety →

Try it on your own lecture

Free: 15 minutes a month, review editor included. No card.

Start free