# Audio Studio Turn a script into speech in a ready-made voice or your own cloned voice, pick the speech model, and use the voice in your AI Influencers and Synora videos. Last updated: 2026-11-05 **Key facts** - Sloosh Audio Studio turns a script into speech: pick a voice, write what it says, pick a speech model and press Generate. - Sloosh Audio Studio's Library has ready-made voices from ElevenLabs, Fish Audio, Google and MiniMax, each with its gender, accent, languages and a sample. - Add my voice in Sloosh Audio Studio clones a voice for free from up to 3 clips (at least 10 seconds of speech in all, each under 20 MB) that you upload or record. - A voice from Sloosh Audio Studio can be given to an [AI Influencer](https://sloosh.ai/docs/ai-influencer.md) and used in [Synora](https://sloosh.ai/docs/synora.md) to make that character talk on camera. - AI tools connected over [MCP](https://sloosh.ai/docs/mcp.md) can list voices, clone one from your uploads and make speech in your Sloosh workspace. Web page: https://sloosh.ai/docs/audio-studio ## What it is Audio Studio is Studio for speech. It has two tabs: Voices, where you pick or add a voice, and My Generations, where every take you made waits with ▶ and a download. Open it from the Studio menu in the top bar, then Audio, or go to [Audio Studio](https://sloosh.ai/studio/audio). ![Audio Studio's Voices tab: voice cards for Roger, Sarah, Laura and Charlie with gender, accent and languages, filter chips for each provider, and the prompt box at the bottom.](https://sloosh.ai/docs/audio-studio-voice-library.webp) ## Make speech 1. On Voices, press a voice card. It goes into the prompt box (the card says In the composer). 2. Write what it says, for example `[excited] Our summer drop is finally here. [whispers] And yes, the linen shirt comes back in sage. [laughs] Don't wait on this one.` 3. Pick a speech model from the chip next to the voice. 4. Set how many takes to make with − and + (up to 4). 5. Press Generate. The button shows the credit cost first. The takes appear in My Generations. ![The prompt box with a script whose [excited], [whispers] and [laughs] tags show as yellow pills, the voice Sarah, the model ElevenLabs v4 (direct) and the Generate button with its cost.](https://sloosh.ai/docs/audio-studio-composer-tags.webp) ### Tags Words in square brackets direct the delivery: `[laughs]`, `[whispers]`, `[excited]`, `[calm]`. They show as pills and are not read aloud. Add tags, above the box, puts them in for you. Which tags work depends on the model; its description in the model list says what it takes. ## Speech models The model decides the sound, the languages and the price. Press the model chip and type to search; each row has a one-line description. ![The speech model list open over the voices, with ElevenLabs Multilingual v2, ElevenLabs v4, ElevenLabs v4 Turbo, Fish Audio Drama 3 and Fish Audio S2.1 Pro.](https://sloosh.ai/docs/audio-studio-models.webp) ## Speech models today | Model | Type | What it does | | --- | --- | --- | | ElevenLabs Music v2.5 | Music | ElevenLabs Music v2.5 text-to-music; full songs or instrumentals from 3 seconds to 1 minute. | | Lyria 3 Clip | Music | Google Lyria 3 Clip short music clips, loops, and previews; takes text and up to 10 image prompts and returns a fixed 30-second MP3 plus lyrics. | | Lyria 3 Pro | Music | Google Lyria 3 Pro flagship music model for full-length songs with verses, choruses, and bridges; takes text and up to 10 image prompts and returns MP3 plus lyrics. | | Lyria 3.5 | Music | Google Lyria 3.5 full-length songs with verses, choruses and bridges; takes text and up to 10 image prompts and returns 44.1 kHz stereo MP3 plus lyrics. | | MiniMax Music 3 | Music | MiniMax Music 3 songs from a style prompt and your lyrics (or instrumental), up to 5 minutes. | | ElevenLabs Sound Effects v2 | Sound effects | ElevenLabs Sound Effects v2; a sound effect from a text description, up to 22 seconds. | | Mirelo SFX 2.0 | Sound effects | Mirelo SFX 2.0 sound effects from a text description. | | Mirelo SFX 2.0 (video) | Sound effects | Mirelo SFX 2.0 scores a video: sound effects synced to the picture, keeping its speech. | | ElevenLabs Multilingual v2 | Speech | ElevenLabs Multilingual v2 text to speech in 29 languages. | | ElevenLabs v4 | Speech | ElevenLabs v4 text to speech; expressive narration with audio tags, up to 5,000 characters. | | ElevenLabs v4 Turbo | Speech | ElevenLabs v4 Turbo text to speech; faster and cheaper v4 voices, up to 5,000 characters. | | Fish Audio Drama 3 | Speech | Fish Audio Drama 3 text to speech you direct like an actor: [directions], sounds, pauses and per-word tags. | | Fish Audio S2.1 Pro | Speech | Fish Audio S2.1 Pro text to speech in 83 languages, with [bracket] directions and your own cloned voices. | | Fish Audio S2.1 Pro Free | Speech | Fish Audio S2.1 Pro text to speech, with the same voices as S2.1 Pro. | | Gemini 3.8 Flash Lite TTS | Speech | Google Gemini 3.8 Flash Lite text to speech; the low-cost tier with the same 30 voices. | | Gemini 3.8 Flash TTS | Speech | Google Gemini 3.8 Flash text to speech with 30 prebuilt voices and separate style instructions. | | MiniMax Speech 2.8 HD | Speech | MiniMax Speech 2.8 HD text to speech; studio-quality narration in over 300 voices, with pause and interjection tags. | | ElevenLabs Voice Changer | Voice change | ElevenLabs speech to speech; re-voices a recording in another voice and keeps its timing and delivery. | | MiniMax H3 Max Lip Sync | Lipsync | MiniMax H3 Max lip sync; animates a portrait image into a talking video as long as the speech, up to 2K. | | Sync Lipsync 2 | Lipsync | Sync Lipsync 2 re-times the mouth in a video to new speech. | | Sync Lipsync 2 Pro | Lipsync | Sync Lipsync 2 Pro; higher-fidelity mouth detail and teeth than Lipsync 2. | | Sync Lipsync 3 | Lipsync | Sync Lipsync 3 (sync-3) re-times a video's lips, face and head to new speech. | | VEED Lipsync v2 | Lipsync | VEED Lipsync v2 re-times the mouth in a video to a new audio track. | | ElevenLabs Dubbing | Dubbing | ElevenLabs dubbing; translates the speech in a video or audio file into another language in the original voices. | | ElevenLabs Scribe v2 | Transcription | ElevenLabs Scribe v2 speech to text with word timings, speaker labels and audio-event tags. | Not every model speaks every voice: a Library voice belongs to its provider, and a cloned voice says which models it works on. ## Clone your voice Add my voice, at the top right, makes a voice from a recording of one person speaking. 1. Press Add my voice. 2. Choose Upload for clips you already have, or Record to read the one-minute script on screen aloud. 3. Give up to 3 clips, at least 10 seconds of speech in all; about a minute clones best. Each file must be under 20 MB. 4. Press Clone and play. Sloosh clones it, which can take a minute or two, and plays it back. The voice joins My voices. ![The Add my voice dialog with Upload and Record, three tips for a good recording, the script to read aloud and the Clone and play button.](https://sloosh.ai/docs/audio-studio-add-my-voice.webp) For the best clone: a quiet room, one voice and no music, spoken the way you would talk on camera. Cloning is free. Only clone your own voice, or a voice whose owner has agreed. ## Give a voice to an AI Influencer On the [AI Influencer](https://sloosh.ai/docs/ai-influencer.md) page, a character without a voice shows + Voice. Press it, pick a voice from My voices or Library (▶ to hear it), and press Add. The voice's name then sits on the character's card, and [Synora](https://sloosh.ai/docs/synora.md) offers it first whenever that character talks. ![Adding a voice to an AI Influencer: the character's ticket at the top, the Library list of voices, and the line Character 13 will speak as Laura with an Add Laura button.](https://sloosh.ai/docs/ai-influencer-add-voice.webp) ## Use it from an AI tool Connect Claude, ChatGPT, Cursor or another AI tool over [MCP](https://sloosh.ai/docs/mcp.md) and it can work with voices for you: 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), and make a voiceover from a script. It quotes the cost before any paid take. **Try asking your AI tool** ```text Write a 20-second voiceover for our linen shirt launch, show me the script, then make it in my cloned voice on Sloosh. ``` ## Cost Generate shows the credits for all the takes before you press it. Speech is priced by the length of the script, so a longer script costs more. Cloning a voice costs nothing. See [Plans and credits](https://sloosh.ai/docs/plans-and-credits.md). ## Frequently asked questions ### What is Sloosh Audio Studio? Audio Studio is Sloosh's speech page: pick a ready-made or cloned voice, write a script, pick a speech model and generate a take you can play and download. ### How do I clone my voice in Sloosh? In [Audio Studio](https://sloosh.ai/studio/audio), press Add my voice, then upload up to 3 clips or record yourself reading the script on screen, at least 10 seconds of speech in all, and press Clone and play. It is free, and the voice joins My voices. ### Which voice models does Sloosh offer? Speech models from ElevenLabs, Fish Audio, Google Gemini and MiniMax. Audio Studio's model chip lists today's models with a line on what each does; this page's Speech models table shows the same list. ### Can I use my cloned voice in a video? Yes. Give it to an AI Influencer with + Voice, or pick it in [Synora](https://sloosh.ai/docs/synora.md) under Voice, and the character says your script on camera in that voice. ### Can I download the audio from Audio Studio? Yes. Every take in My Generations has a download button; hover a card on a computer to see it. ### Whose voice can I clone in Sloosh? Your own, or someone who has agreed to have an AI voice made of them. Never clone a child or a public figure.