Gemini Omni Profile: AI Tool

Create a named voice profile with custom timbre, style, and emotion. The returned voice ID can be used in Gemini Omni video generation to assign a consistent voice character to your videos.

📝

Overview

About this model

Create a named voice profile by selecting a preset voice and adding a custom description of its timbre, style, and emotion. The resulting voice ID can be attached to Gemini Omni video generations to give characters a consistent, reusable voice across multiple videos.

1Character Consistency: Assign the same voice to a recurring character across multiple generated videos.
2Narrative Branding: Build a library of named voice profiles for different brand personas or show characters.
3Localization Prep: Register distinct voice styles for different markets before bulk video production.
💰

Pricing & Value

Cost analysis

muapiappFree

No charge for voice profile creation.

Fal.aiNot available

Gemini Omni Audio profile creation is not offered.

ReplicateNot available

Gemini Omni Audio profile creation is not offered.

** Competitor pricing is estimated based on similar model architectures and usage tiers.

⚙️

Technical Details

Configuration schema

Base VoiceEnum (30 options)

Preset voice to use as the base for this profile.

Default Value-
Profile Namestring

Name for this voice profile. Maximum 210 characters.

Default ValueMy Custom Voice
Voice Descriptionstring

Describe the voice timbre, style, and emotion. Maximum 20,000 characters.

Default ValueWarm, calm tone with a slight British accent and gentle pacing.
Example Dialoguestring

A short sample sentence the voice would say. Maximum 120 characters.

Default ValueHello, I am here to guide you through today's journey.
📖

Implementation Guide

Developer documentation

How to Use Gemini Omni Audio

  1. Pick a base voice: Choose one of the 30 preset voices (e.g., achernar, charon, zephyr) that best matches the vocal character you want.

  2. Name your profile: Give it a memorable name (up to 210 characters) — this name is returned with the ID for easy reference later.

  3. Describe the voice (optional): Add a voice_description (up to 20,000 characters) specifying timbre, pacing, accent, or emotional tone.

  4. Add example dialogue (optional): Provide a short sample sentence (up to 120 characters) to help anchor the voice style.

  5. Submit and save the ID: The response returns an audio_id. Store it — you will pass this ID in the audio_ids field when using Gemini Omni video generation endpoints.

Common Questions

Frequently asked

What is the output of this endpoint?

An `audio_id` string and the profile name. This ID is not a playable audio file — it is a voice configuration token you pass to Gemini Omni video endpoints.

How do I use the voice ID in a video?

Pass the `audio_id` value in the `audio_ids` field when calling the Gemini Omni text-to-video, image-to-video, or video-edit endpoints.

Can I create multiple voice profiles?

Yes — each call creates one profile and returns a unique ID. You can build a library of profiles and reuse them across different video projects.