# Introducing Audio Studio: give your character a consistent voice Audio Studio turns a script into speech with voices from ElevenLabs, Fish Audio, Google and MiniMax, clones your own voice for free, and gives that voice to your AI Influencers and your agents. 5 Nov 2026 · 4 min read · Product · By Sloosh Web page: https://sloosh.ai/blog/introducing-audio-studio **Write a script, pick a voice and press Generate.** Audio Studio is Studio for speech: the best voice models in one place, your own cloned voice in about a minute, and a voice your characters keep in every video. [Video: The voice Sarah reading a script with [excited], [whispers] and [laughs] tags, made in Audio Studio.](https://cdn.pro.shopos.ai/sloosh/o/93664a0d00debd5d/01a1abe4-66b0-7d31-858a-cfe50de72900.mp4) _Made in Audio Studio: the voice Sarah reading a script with [excited], [whispers] and [laughs] tags._ ## TL;DR - **Many voice models, one page:** speech models from ElevenLabs, Fish Audio, Google Gemini and MiniMax, with a library of ready-made voices. - **Direct the delivery:** tags like `[laughs]`, `[whispers]` and `[excited]` set how each line is said. - **Clone your voice for free:** upload or record about a minute of speech and press Clone and play. - **Give it to a character:** attach any voice to an [AI Influencer](https://sloosh.ai/ai-influencer), and make them talk on camera with [Synora](https://sloosh.ai/blog/introducing-synora). - **Use it from your agent:** Claude, ChatGPT, Cursor and other AI tools can list voices, clone and make voiceovers over [Sloosh MCP](https://sloosh.ai/mcp). ## Why a voice matters Your character can look the same in every video and still feel like a stranger if the voice changes. A consistent voice is what makes a face into a person your audience recognises. Audio Studio makes the voice a saved part of your workspace, like a character or a brand kit, instead of something you hunt for each time. ## Pick from the best voice models Open [**Audio Studio**](https://sloosh.ai/studio/audio) from the Studio menu. The **Voices** tab shows the Library: each card has the voice's gender, accent, languages and a sample. Filter by provider or search by name, then press a card to put that voice in the prompt box. ![Audio Studio's Voices tab with voice cards for Roger, Sarah, Laura and Charlie, filter chips for each provider and the prompt box.](https://cdn.pro.shopos.ai/sloosh/o/93664a0d00debd5d/01a1abd9-4d55-72df-bb17-b6180c294082.webp) Then pick the speech model. Each one sounds different and suits a different job: ![The speech model list: ElevenLabs Multilingual v2, ElevenLabs v4, ElevenLabs v4 Turbo, Fish Audio Drama 3 and Fish Audio S2.1 Pro.](https://cdn.pro.shopos.ai/sloosh/o/93664a0d00debd5d/01a1abd9-45b1-734e-a93e-95e4f827a7b0.webp) | If you need | Try | | --- | --- | | Expressive narration with audio tags | ElevenLabs v4 | | A faster, cheaper take | ElevenLabs v4 Turbo | | Many languages | ElevenLabs Multilingual v2, Fish Audio S2.1 Pro | | Acting: directions, sounds and pauses | Fish Audio Drama 3 | | Your own cloned voice | Fish Audio S2.1 Pro (your voice's card lists the models it works on) | | Low cost at scale | Gemini 3.8 Flash Lite TTS | | Hundreds of stock voices | MiniMax Speech 2.8 HD | _The model list grows; Audio Studio and the _[_Audio Studio docs_](https://sloosh.ai/docs/audio-studio)_ always show today's models._ ## Direct the delivery with tags Words in square brackets tell the voice how to say a line. They show as yellow pills and are not read aloud. Press **Add tags** and Sloosh adds them for you. Set how many takes you want (up to 4) and press **Generate**. The button shows the credit cost first. ![The prompt box with [excited], [whispers] and [laughs] tags as yellow pills, the voice Sarah, the model ElevenLabs v4 (direct) and the Generate button with its cost.](https://cdn.pro.shopos.ai/sloosh/o/93664a0d00debd5d/01a1abd9-3ce2-71ab-aeb0-56e960e20adf.webp) > **Tip: Try this script.** "[excited] Our summer drop is finally here. [whispers] And yes, the linen shirt comes back in sage. [laughs] Don't wait on this one." Every take waits in **My Generations** with play and download. ![My Generations in Audio Studio: speech takes with their voice, length, model and cost, and the new Sarah take with its tagged script.](https://cdn.pro.shopos.ai/sloosh/o/93664a0d00debd5d/01a1abd9-4311-74f7-884a-ce62ede69a5f.webp) ## Clone your own voice in a minute Press **Add my voice**. Choose **Upload** for clips you already have, or **Record** to read the on-screen script aloud. Up to 3 clips and at least 10 seconds of speech in all; about a minute clones best. Press **Clone and play** and your voice joins **My voices**. Cloning is free. ![The Add my voice dialog: Upload or Record, three tips for a clean recording, the script to read aloud and the Clone and play button.](https://cdn.pro.shopos.ai/sloosh/o/93664a0d00debd5d/01a1abd9-39c8-7fa5-9045-2f97ca8ab48b.webp) **Tips for a great clone:** 1. A quiet room, one voice and no music. 2. Speak as you would on camera: same pace, same energy. 3. A minute or more of you talking. ## Attach a voice to your AI Influencer On the [**AI Influencer**](https://sloosh.ai/ai-influencer) page, any character without a voice shows **+ Voice**. Pick a voice from **My voices** or the **Library**, press ▶ to hear it, and press **Add**. From then on that character sounds the same everywhere, and [Synora](https://sloosh.ai/blog/introducing-synora) uses that voice first whenever they talk on camera. ![Adding a voice to an AI Influencer: the character's ticket, the Library list of voices, and "Character 13 will speak as Laura" with an Add Laura button.](https://cdn.pro.shopos.ai/sloosh/o/93664a0d00debd5d/01a1abd9-350c-7cec-840a-81c98b51d96c.webp) Upload or create a character, add speech, and bring them to life with lip sync. That's the whole loop: **AI Influencer** for the face, **Audio Studio** for the voice, **Synora** for the video. ## Use voices from your AI agent over MCP Connect Sloosh to Claude, ChatGPT, Cursor, VS Code, Codex and more from the [**MCP page**](https://sloosh.ai/mcp). Your agent can then: - list your workspace's voices and the Library, - clone a voice from audio you uploaded (it asks you to confirm you are the speaker, or have their permission), - write a script with you and make the voiceover in the voice you choose, quoting the cost first. ![The Sloosh MCP page: "Sloosh works with your agent." with Claude, ChatGPT, Claude Code, Codex, Cursor, VS Code, Gemini CLI, Windsurf and LM Studio.](https://cdn.pro.shopos.ai/sloosh/o/93664a0d00debd5d/01a1abd9-4fe4-711c-8403-574b15a2a446.webp) > **Tip: Ask your agent.** "Write a 20-second voiceover for our linen shirt launch, show me the script, then make it in my cloned voice on Sloosh." ## Use it responsibly Only clone your own voice, or the voice of someone who has agreed to it. Never clone a child or a public figure. Say that the voice was made with AI where the platform asks you to. ## Try it Open [**Audio Studio**](https://sloosh.ai/studio/audio), pick a voice and generate your first take. The [**Audio Studio docs**](https://sloosh.ai/docs/audio-studio) cover every model, tag and setting. ## FAQ **What is Sloosh Audio Studio?** Audio Studio is Sloosh's text-to-speech page. You pick a ready-made or cloned voice, write a script, pick a speech model and generate takes you can play and download. **Which voice models are in Audio Studio?** Speech models from ElevenLabs (including v4 and Multilingual v2), Fish Audio (S2.1 Pro and Drama 3), Google Gemini TTS and MiniMax Speech. The model chip in Audio Studio always shows today's list. **How do I clone my voice?** Press Add my voice, upload up to 3 clips or record yourself reading the on-screen script (at least 10 seconds of speech in all, about a minute is best), and press Clone and play. It's free. **Can I use my cloned voice in a video?** Yes. Give it to an AI Influencer with + Voice, or pick it in Synora under Voice. The character then says your script on camera in that voice, with lips in sync. **How do I make an AI voice laugh, whisper or sound excited?** Add tags in square brackets, like [laughs], [whispers] or [excited], before the words they apply to, or press Add tags. Which tags work depends on the model. **Can I download the audio?** Yes. Every take in My Generations has a download button. **Can my AI agent make voiceovers in Sloosh?** Yes. Connect it over Sloosh MCP and it can list voices, clone a voice from your uploads and make speech, within the daily credit limit you set. **How much does it cost?** Speech uses Sloosh credits and is priced by the length of the script; Generate shows the cost first. Cloning a voice is free. See pricing.