OmniVoice

Generate natural voiceovers and cloned voices.

OmniVoice covers fast text narration and reference-based voice cloning with simple speed and voice-style controls.

Trait-based voices
Reference cloning
Speed control

What OmniVoice is good for

Composable voice traits

Use validated descriptors such as accent, age, gender, pitch, and whisper style.

Speed control

Adjust delivery pace from slow narration to faster social-ready voiceovers.

Clone from audio

Use a clean sample to produce speech in the reference voice.

Text narration

Turn text into a natural spoken track.

Describe voice traits, set the pacing, and generate audio from plain text.

0/3000
SlowNormalFast
Use cases

Built for production voice workflows.

Use these pages as a focused audio workspace for scripts, product media, education content, and repeatable brand narration.

Product videos

Generate clear narration for feature launches, onboarding clips, release notes, and short demo videos.

Creator voiceovers

Produce consistent narration for reels, explainers, tutorials, and channel updates without recording every take.

Localized narration

Prepare alternate voice directions for region-specific campaigns, ads, and help center content.

Reference voice reuse

Use voice cloning when a project needs the same speaking identity across multiple scripts.

Workflow

A simple path from script to usable audio.

The generation panel handles task creation, status polling, preview, and download. The supporting content helps you prepare better inputs before submitting.

1

Choose the mode

Use Text to Speech for new narration, or Voice Clone when a reference speaker should guide the output.

2

Prepare the input

Paste the final script, add voice descriptors, and upload clean reference audio for cloning tasks.

3

Generate and review

Submit the task, preview the finished audio, then download the result for editing or publishing.

Mode comparison

ModeBest forInputsOutput
Text to SpeechFast narration and voiceover draftsText, voice description, speedGenerated speech audio
Voice CloneKeeping a known voice across multiple scriptsText, reference audio, optional transcript, speedCloned speech audio

Input guide

Voice description

Combine age, accent, gender, pitch, and delivery traits with commas, such as female, young adult, moderate pitch.

Reference audio

Use a clean clip with one speaker, low background noise, and a speaking style close to the desired result.

Script length

Short paragraphs are easier to review and regenerate. Split long scripts into scene-level sections.

Examples

Copy-ready prompts for stronger first results.

Use these examples as starting points, then adapt them to your project, audience, and delivery style.

Voice description

female, young adult, american accent, moderate pitch, calm studio narration

Product script

Create a polished 20-second product update with a steady pace and confident delivery.

Clone reference text

Paste the transcript of the uploaded reference clip when available to improve pronunciation alignment.

FAQ

Common questions before generating.

A short reference for choosing the right mode and preparing inputs before sending a task to the backend.

When should I use OmniVoice instead of Qwen3 TTS?

Use OmniVoice when you want simple voice trait controls, speed adjustment, or a fast reference-clone workflow.

What kind of audio works best for cloning?

Use a clear single-speaker clip with minimal background sound. A matching transcript can help the backend model understand the reference more accurately.

Can I change the generated speaking speed?

Yes. OmniVoice modes include speed control, so you can generate slower narration or faster delivery for short-form content.