AI Voice Remixing

Start with a voice you like. Describe the changes you want, listen to the result, and save a new voice for your next story, video or conversation.

0/600 characters

15,000 credits

Your voice library

Voice remixing: before and after

Compare the original and remixed voices reading the same short passage. Use a prompt as a starting point, then adjust it for your own project. Each new generation can sound different.

A warmer storyteller

Lower pitch · Warmth · Audiobooks

Original · Puck

Remixed

View remix prompt

Keep the friendly personality and American English accent. Lower the pitch slightly and add a warm, gently textured tone suited to a thoughtful storyteller.

A softer guide

Soft tone · Calm pace · Meditation

Original · Zephyr

Remixed

View remix prompt

Keep the clear pronunciation and American English accent. Soften the tone and reduce the vocal energy for a calm, unhurried meditation guide.

A British narrator

British English · Clear delivery · Courses

Original · Kore

Remixed

View remix prompt

Change the accent to British English. Keep the confident delivery, clear articulation and overall vocal character, with a relaxed educational narration style.

An energetic host

More energy · Mature character · Introductions

Original · Gacrux

Remixed

View remix prompt

Keep the mature vocal character. Add brighter energy and a more animated, welcoming delivery for a podcast introduction, without shouting.

How to remix a voice

Build on a sound that already works, then make a focused change.

  1. 01

    Choose a base voice

    Listen to the presets, or choose a voice you’ve saved with Voice Design or Voice Remixing. Pick the voice closest to what you need.

  2. 02

    Describe the change

    Say what you want to keep and what should change. Ask for a lower pitch, a warmer tone, a younger or older voice, or a different accent. Start with one or two traits, then generate your new voice preview.

  3. 03

    Listen, save and create

    Listen to the preview, then try the voice with your own text. Name and save the version you like, then use it in single-speaker or multi-speaker text to speech. Save a result before using it as the base for another remix.

Voice remixing prompts that give clear direction

Describe a change to the selected voice, rather than writing a complete new character biography.

Keep the qualities you like

Name the traits that should remain: the current accent, a friendly tone or clear pronunciation. Then describe your adjustment. For a brand narrator, you might keep the approachable tone while adding a more mature sound. Bigger changes may make it sound like a different person, so listen to each preview.

Make an AI voice deeper or softer

Try “slightly deeper, with the same accent” or “softer and more relaxed.” Test a passage with questions, short sentences and longer phrases to judge whether the result fits your narration.

Change an accent with context

Remixing keeps the base voice's language and accent by default. To change them, state the target in your prompt, such as British English, and describe which other traits to preserve. Check names and difficult words with your own script before committing to a long recording.

Separate the voice from the performance

Remix a voice when you want a reusable change to its sound. Use emotion tags in the text-to-speech editor to change how a particular line is spoken. A character can keep the same voice while sounding excited in one scene and reflective in the next.

Choose the right starting point

These tools work with different inputs. Pick the one that matches what you already have.

Voice Remixing

You already like an AI voice and want to adjust it. Choose a preset, a saved Voice Design voice, or a saved remix. Describe the changes to create a new voice you can reuse.

Remix a voice

Voice Design

You have a character in mind but no starting voice. Describe the speaker from scratch, generate a preview and save the voice. You can later refine a saved design here.

Design a new voice

Voice Cloning

Upload a recording to create a voice based on the speaker. Cloned voices cannot be used as a base for Voice Remixing.

Explore voice cloning

Voice Changer

You have a recording and want to transform the spoken audio into a different voice. Start with the voice changer when the recording itself is your input.

Change a recording’s voice

Questions about voice remixing

Answers about supported voices, accents, credits and saving.

AI voice remixing lets you change an existing AI voice using a text prompt. You can make it sound deeper, warmer, older or more energetic, or ask for a different accent. Each result is a new voice, so your original stays unchanged.


You can choose from 30 preset voices, voices you’ve saved with Voice Design, and your saved remixes. Custom voices must still be available in your own FlowSpeech library. Save a new preview before using it as the base for another remix.


No. Voice Remixing uses the voices in the Base voice menu. It does not accept audio uploads or cloned voices. To create a voice from a recording, use Voice Cloning.

Create a voice from a recording

Yes. Describe how you want it to sound—for example, “slightly deeper, but keep the same accent.” You can also ask for a younger or older voice. Results vary, so listen to the preview and adjust your prompt if needed.


No. A remix keeps the base voice’s language and accent by default. If you want to change them, say so in your prompt—for example, “Give this voice a British accent.” Listen to the preview to check the result.


No. Each remix is a separate voice, so the original stays unchanged. Once the remix is ready, deleting its base voice won’t delete the remix. Saved remixes use Voice Remixing slots. Voice Design has its own saved voice limit, so saving a remix does not use a Voice Design slot.


Yes. Choose Try Text to Speech to test it with your own script. Save the voice to use it in the main text-to-speech editor, including conversations with multiple speakers. Generating speech uses credits separately from creating the voice.


Each new remix costs 15,000 credits, including an audio preview. New speech costs 1 credit per billable character. Replaying existing audio and saving a voice are free. Without an active subscription when creation finishes, a voice is available for 72 hours; saving does not extend that time. Plan limits apply separately to saved Voice Remixing voices. If we confirm that generation failed, we return the credits for that remix.

Compare plans and voice limits

Without an active subscription when creation finishes, a voice is available for 72 hours, whether saved or not. Subscribe or renew before it expires to keep using it under your plan’s saved-voice rules. Voices created during an active subscription keep the existing rules: unsaved previews last one hour; saved voices last up to one year from creation. Saving is free and never restarts the expiry date. Voice Design and Voice Remixing have separate saved voice limits and two preview slots each, including pending requests.


Choose a voice to remix

Start Remixing