AI dubbing for broadcasters
Your archive, in another language.
Tolkr™ dubs recorded programmes and follows foreign channels on your organisation's own dedicated systems. Each sentence starts with its original, each speaker has a voice, and your team receives everything in one folder.
What your organisation gains.
- Your archive reaches new audiences.
- Programmes that never justified a studio dub can be watched in another language.
- Foreign stories reach your desk sooner.
- Follow a channel or a press conference in your language while it happens, and quote it when it ends.
- Unreleased material stays with you.
- The work happens on your own systems, so a programme does not have to go out to a dubbing service.
- Your people stay in charge of the result.
- Editors can check every sentence, and the sound team can have the separate tracks.
- You judge it on your own programme.
- Bring a short clip and we run Tolkr™ on it with you, before anyone commits to anything.
- Content
- More of the catalogue, in more languages
- News
- What a foreign channel says, as it says it
- Post production
- A finished dub, with the parts to refine it
- IT and security
- Your own systems, nothing sent during use
Give it a programme. Receive a finished dub.
Nobody has to run this step by step. Tolkr™ works through a programme by itself and shows where it is.
Choose
Pick the file and the language.
Let it work
Tolkr™ finds the speakers, translates and voices each sentence.
Check and hand on
Listen, then pass the files to your team.
Recorded from Tolkr™ Studio. If the workstation restarts, start the dub again and it continues from where it stopped.
Each sentence starts with its original.
A dubbed voice that runs long slides away from the picture. Tolkr™ places every sentence where the original sentence starts.
- Start
- With the original sentence
- End
- Before the next sentence begins
- Original sound
- Lowered under the voice, and kept as its own track
- Picture
- Copied, never encoded again
Recorded from Tolkr™ Studio. Every sentence is listed with its time, what was said and how it was translated, so an editor can check it without watching the whole programme.
Mixed to -23 LUFS.
A dub master is delivered to set values, with the original under the voice. Your sound engineer can still mix again from the separate tracks.
- Loudness
- -23 LUFS
- True peak, not above
- -1 dBTP
- Sample rate
- 48 kHz
- Bit depth
- 24 bit
On dub masters. The MP4 sound track is AAC.
Tolkr™ Studio reports the result of each dub.
You see whether every sentence starts with its original, and what the finished mix measures, before the file leaves your desk.
Follow a channel as it runs.
The channel plays on the PC. Tolkr™ shows both lines as captions, speaks the translation as the speaker talks and saves the session as a bilingual transcript.
- Source
- What the PC plays, one app, a microphone or a file
- Captions
- Both lines, in a window that floats over any program
- Voice
- A natural voice, or captions alone
- Afterwards
- A transcript with its sound, to read and quote from
Vragen van de pers volgen daarna.
Questions from the press follow afterwards.
Try the sliders, there is no sound on this page.
One dub. A folder for the whole team.
Each person finds their part, without dubbing again.
- Editor
- The dubbed video and the subtitles
- Sound engineer
- The mix, the voice alone and the original under it
- Subtitler
- Both subtitle files and the sentence list
- Archive
- The dubbed video and a text record of it
- Video
- MP4, MOV
- Sound
- MP3, WAV, FLAC, M4A, Opus
- Subtitles
- SRT, VTT
Need another format later? Save the files again, without dubbing again.
Set up with you. On every workstation that needs it.
We install Tolkr™ with you on each workstation, with the languages, voices and presets your team works with. Each one then runs by itself.
- Listen
- Understand
- Translate
- Speak
- Listen
- Understand
- Translate
- Speak
- Listen
- Understand
- Translate
- Speak
- Platform
- Windows 11
- Hardware
- Faster with a graphics card, and it also runs on the processor. Tolkr™ sets itself up for the workstation.
- Files
- Ordinary files in a folder you control
- Network
- During use, it makes no connection outside the PC.
What a head of post asks first.
Not answered here? Ask the people who build it
How long does a dub take?
It depends on the workstation and the programme, so we run it on your own material in a demo. A card shows the time left while it works.
Can the sound engineer mix again?
Yes. The voice and the original under it come as separate files when you turn that choice on.
Does it tell speakers apart?
Yes. Tolkr™ tells the speakers apart and gives each one a voice. It can suggest names from the conversation, and subtitles can carry them.
What if a workstation restarts during a long programme?
Start the same dub again and it continues from where it stopped.
How does it follow a channel?
The channel plays on the PC, in the browser or a player, and Tolkr™ follows it with captions and a voice.
Can it dub long programmes?
Yes. A card shows the time left, and a dub that was stopped continues where it stopped.
Which languages are covered?
Captions in 130 languages, 70 understood as they are spoken, and a natural voice in 36. See the language list.
Shall we listen?
Bring a short clip or a channel you need to follow. We run Tolkr™ on it with you.
Request a demo