Multi-speaker AI dubbing
Every speaker, a voice of their own.
In a dubbed file, Tolkr™ tells the speakers apart, gives each a voice of their own and can suggest their names. Every sentence knows who speaks, and to whom.
Recorded from Tolkr™ Studio. A panel of three, 
Spanish to 
English.
Who speaks when
One programme, every voice told apart.
A panel discussion with three speakers. Tolkr™ tells them apart and gives each a voice of their own.
- Told apart
- In every dub, by itself
- Reported
- The dub says how many speakers it found
- A voice each
- Man or woman from the voice, yours to change
- Dub or keep
- For each speaker, your choice
- Quiet voices
- Kept, even when they speak softly
Step by step
Find, voice, name, decide.
Telling the speakers apart is part of every dub. The rest wait for you in the Speakers box, and each can be changed later.
- 01Find
Tolkr™ tells the speakers apart, and the dub shows who it found.
- 02Voice
Each speaker gets a voice of their own. Hear a sample of each one.
- 03Name
Type a name, or let Tolkr™ suggest one from what is said.
- 04Decide
Dub each speaker, or keep their original voice.
- Same personJoin two
Choose Same person as, and two speakers share one voice.
- Any timeChange later
Every choice in the Speakers box can be changed after the dub.
- Your switchOn or off
Each part of the speaker features can be switched on or off.
Names
Names from what is said.
People introduce themselves and call each other by name. Tolkr™ suggests one name for each speaker, from what is said.
What is said
Speaker 1Good evening, I'm Dana, and this is the harbour panel.
Speaker 1Marco, how did the new boats do this summer?
What Tolkr™ suggests
Introduced herselfSpeaker 1 is Dana
Called by nameSpeaker 2 is Marco
- One eachFrom every line
Each speaker gets one name, from what is said.
- Two waysIntroduced or called
A speaker who gives their own name, or is called by it, gets that name.
- Your checkBefore you publish
Names are suggestions. Check them in the Speakers box before the dub goes out.
Who speaks to whom
Man or woman, in every sentence.
In many languages a word changes with who speaks and who is spoken to. Tolkr™ knows both for every sentence, so the dub says it the way a native speaker would.
English→SpanishDana, a woman, to Marco
Are you ready?¿Estás listo?
I was so worried.Estaba tan preocupada.
Marco, a man, to Dana
Are you ready?¿Estás lista?
I was so worried.Estaba tan preocupado.
- Who speaksMan or woman
The dub follows whether the speaker is a man or a woman.
- To whomThe listener
And whether the person spoken to is a man or a woman.
- NativeAs spoken
So the dub says it the way a native speaker would.
Where it shows
Found once. Used everywhere.
The speakers of a dub carry into every file it delivers.
- The dub
- A voice per speaker, each sentence at its original
- Subtitles
- Lines can carry the speaker's name
- Sentence list
- Speaker, man or woman, said to and voice, for every sentence
- Speakers box
- Who was found, with a sample of each
- Works in
- Dub a file
- Runs on
- Your own systems
- Each part
- Switched on or off in the settings
Your questions
Asked by your role.
Choose your role. Each one has the questions people in it ask first, answered.
Editor3 questions
- Does it tell the speakers apart?
Yes. Each speaker gets a voice of their own, and you see them all in the Speakers box.
- Can I change a voice?
Yes. Tolkr™ hears whether it is a man or a woman, and you can change it in the Speakers box.
- Can I keep one speaker in the original?
Yes. Choose for each speaker whether to dub them.
Subtitler3 questions
- Can lines carry the speaker's name?
Yes. Subtitle lines can carry the speaker's name, if you choose.
- Should I check suggested names?
Yes. Names are suggestions from what is said. Check them before you publish.
- What does the sentence list hold?
For every sentence: its times, both texts, the speaker, man or woman, who it is said to and the voice.
Head of IT3 questions
- Does it need the internet to translate?
No. Recognition, translation and the voice run on your own systems.
- What does it run on?
Windows 11. A graphics card makes it faster, and it also runs on the processor. Tolkr™ sets itself up for each workstation.
- Can each part be switched off?
Yes. Each part of the speaker features can be switched on or off.
Hear your programme, speaker by speaker.
We dub a short clip of yours in a demo, so you hear each voice yourself.
Request a demo