WORDS IN. YOUR SOUND OUT.

Voice Over Studio.

Create a cast. Direct every scene. Make the final mix yours.

Open project

01 / Script & cast

0 characters

Project name

English voices · 12,000 characters per project · 40 scenes maximum. Each scene can have its own speaker, pace and pause.

Pronunciation dictionary

One replacement per line: original = spoken form

Literal, case-sensitive replacements apply to speech; captions keep your original words.

Ready. First generation downloads an approximately 90 MB model plus runtime files.

03 / Listen & export

No render yet

Generate a script or import voice audio to begin.

A complete local narration workflow.

Write or import a script, assign US or British English voices, then generate your scenes. Adjust the mix without regenerating speech. Import your own recordings for the same sound controls. Save a portable project with source audio, or download a 16-bit WAV and scene-timed SRT captions for your video editor.

Neural speech uses Kokoro.js. Model and engine files download from Hugging Face and jsDelivr; script and audio processing happen on your device. Generation speed depends on your device. This tool does not clone voices, guarantee emotional delivery, or provide word-aligned subtitles.

FREE TO USE / NO ADS / NO LOGIN REQUIRED

Create a voice over with multiple AI voices

Generate multi-speaker English narration locally, direct each scene and download a finished WAV mix.

Choose from 28 US and British English neural voices
Assign a voice, speaking pace and pause to each scene
Import recordings, trim audio and apply pitch, EQ, compression and echo
Mix background music with speech ducking and peak normalization
Download WAV audio, scene-level SRT captions and portable projects

Write or import text, assign speakers and generate speech. Use sound direction to shape the mix, listen to the result and export WAV. First use downloads the speech model and runtime. Pitch modulation also changes playback duration.

Popular ways to use Voice Over Studio

Create a two-speaker voice over for a video or presentation.
Produce podcast narration with background music and scene captions.
Change the pitch and sound of an imported voice recording.

Voice Over Studio questions

Can I download the generated voice over?

Yes. Export a 16-bit PCM WAV in mono or stereo at 24, 44.1 or 48 kHz, and scene-level SRT captions.

Does this tool clone voices or support every language?

No. It provides 28 preset English voices with US and British accents. It does not clone voices or guarantee emotional acting.

Are my scripts uploaded?

No. Speech generation and audio mixing run locally. Engine and model files download on first use.

How processing and privacy work

Scripts and audio are processed on your device. The speech engine and approximately 90 MB model download from jsDelivr and Hugging Face. An internet connection is needed for uncached assets.