Who speaks, and when
Speaker turns are detected so every voice stays attached to the right person.
Your podcast,
with every voice.
MadeinVox detects who is speaking and when. Every host and every guest keeps a consistent voice, episode after episode.
H
Host
Voice: Adam
G
Guest
Voice: Sarah
- consistent per speaker
- 1 voice
- consistent per speaker
- voices in the library
- 18,000+
- voices in the library
- dubbed video ready to publish
- MP4
- dubbed video ready to publish
From source video to dubbed version, in four steps.
- 1
Upload the episode
Video or YouTube link: the transcript separates the speakers automatically.
- 2
Assign the voices
Listen to the samples and give each person a voice.
- 3
Review the conversation
Guest names, references and follow-ups can be corrected in the editor.
- 4
Export
Get the dubbed episode back with the same conversational rhythm.
A cast that doesn’t change
The same host keeps the same voice from one episode to the next.
Names pronounced right
Correct a guest’s or a brand’s name once, before generation.
Long formats too
An hour-long interview is reviewed segment by segment, without losing the thread.
Intro and jingles intact
Sound branding and music stay under the new voices.
Meaning, tone, rhythm.
Every line is translated to be spoken, not just read, then reviewed before the voice is generated.
« Honnêtement, je ne pensais pas que ça marcherait aussi vite. »
“Sinceramente, no pensaba que funcionaría tan rápido.”
Dubbed version · Spanish
Frequently asked questions
- Do guests get different voices?
- Yes. Every detected speaker gets their own voice, which you can change before generation.
- My podcast is in Darija. Is that possible?
- Yes, Moroccan Darija is a full language in the studio, as a source and as a target.
- Can I redo a single line?
- Yes. After a correction, only the edited lines are regenerated.
Other use cases
MadeinVox
Your next video, in a new language.
Create your account: your first 5 minutes of dubbing are free.