AI Voice Re-Dubbing for Any Recording

A free, browser-based re-dubbing tool. Upload a voice recording or video, get an automatic transcript aligned to the original timing, then re-voice it with a natural AI voice. Assign different voices to different characters, re-dub a single speaker while keeping the others, or just type text and have it spoken — with correct Russian stress added automatically.

How to re-voice a recording

Upload an audio or video file by dropping it or clicking browse

Pick the spoken language and the new voice you want

Click "Transcribe & re-voice" — the tool transcribes the speech and re-voices each line, fitting it to the original timing

Edit any line of the transcript, change its voice, or mark a speaker as "keep original", then re-render

Play the result and download the re-dubbed audio (or video)

Replace or switch the voices in a recording, in your browser

Spoken language

Voice engine

New voice

Transcript quality

Speakers

Keep background music

Drop an audio or video file here or browse

MP3, WAV, M4A, OGG, FLAC, MP4, WebM and more

MP3, WAV, M4A, OGG, FLAC, MP4, WebM ·

Original

Re-dubbed

Transcript — edit text, pick a voice per line

Russian stress is added automatically. For names or ambiguous words, type + before the stressed vowel: g+olosa.

Voices: Supertonic 3 (OpenRAIL-M) and Kokoro (Apache 2.0). Russian stress from OpenRussian / Wiktionary (CC-BY-SA 4.0). Speech recognition by Whisper, vocal separation by Kim Vocal 2, time-stretch by Signalsmith Stretch (MIT). Please do not use this tool to impersonate real people or to create misleading or unlawful content.

Features

Re-voice any recording: upload audio or video and replace the spoken voice while keeping the original timing and pace Multiple voices per file — assign a different voice to each character and switch line by line Re-dub one speaker only — leave the other characters as the original audio Natural neural voices in 30+ languages, with correct Russian stress added from a 535,000-word dictionary Character voice presets (e.g. a stern governess, a cheeky kid) derived by pitch and formant shaping — original voices, no impersonation Runs in your browser via WebAssembly and WebGPU — works offline after the first load

Why use this re-dub tool

Free and unlimited, no signup. Keeps the original timing so a re-dubbed clip stays in sync. Multiple voices and selective per-speaker re-dubbing in one place. Correct Russian stress out of the box. Original character presets instead of risky celebrity impersonation. Open, browser-based processing with honest, clean-licensed models.

Frequently Asked Questions

Can I make a voice sound like a specific actor or cartoon character?

No — cloning a real, recognizable person's voice is not offered, for legal and ethical reasons. Instead, the tool provides original character-style presets (a stern older woman, a cheeky child, a deep narrator) created by reshaping a clean AI voice. They evoke an archetype without imitating any real person.

Can I use two different voices in one file?

Yes. After transcription each line can be assigned its own voice, so a dialogue can use one voice for character A and another for character B. You can switch the voice on any individual line.

Can I re-dub only one character and keep the other?

Yes. Mark the lines you want to keep as "keep original" — those segments use the untouched original audio, while the rest are re-voiced. This works cleanly when the speakers do not talk over each other.

Is the Russian stress correct?

Common words are stressed automatically from a 535,000-form dictionary. Homographs (за́мок / замо́к) and proper names can be ambiguous — for those, type a + before the stressed vowel and it will be read that way.

Why does the first run take longer?

The voice model is downloaded the first time you use the HD engine, then cached by your browser for later visits. You can also use the instant system voice, which needs no download.

What languages are supported?

The HD neural engine supports Russian and 30 other languages. For languages it does not cover, the tool falls back to your device's built-in system voices.

💡 Want us to improve this tool just for you?

We can — and it's free! Just send us a quick message with your idea. If you'd like to discuss it in detail, leave your email and we'll get back to you. You can stay anonymous.

How do you rate this tool?

Thank you for your rating!

Change my rating

Want to leave a comment?

Want to share more? Leave a comment!

Thank you! Your comment will appear after moderation.

Who is this tool for?

Published 04 Jun 2026 Updated 05 Jun 2026