Vocalune
Coming soon
Text to speech for Mac
Paste an article, a chapter or a script and hear it read aloud in a natural AI voice. The models run on your Mac, so your text stays private and everything works offline.
What you can do
Vocalune is built for reading long text aloud, making voiceovers, and listening to articles and documents. Here is what each part of the app does.
Studio
Type, paste or open a document. Vocalune reads plain text, Markdown, RTF, Word (.docx), HTML and PDF files. Choose a voice, set the speed from 0.5x to 2x, and press Speak.
Voices
Listen to any voice before you use it, sort by language and voice type, and star your favorites. With Pro you can mix two or three Kokoro voices into a new one, such as a warm narrator with a deeper tone.
Voice design and voice cloning (Pro)
Write a description such as "a calm older man with a low, warm voice". Qwen3-TTS Voice Design or VoxCPM2 creates a matching voice in any of their languages. Good for characters, games and prototypes.
Record or import about ten seconds of speech and the voice cloning models read any text in that voice. You confirm that it is your own voice or that you have the speaker's permission, and the voice is saved only on your Mac.
The larger Qwen3-TTS models and VoxCPM2 follow style directions such as "speak softly" or "sound excited", so the same voice can fit a bedtime story or a product video.
Library
Each clip goes to the Library with its text, voice, model and speed. Search by text or voice, mark favorites, or open a clip in the Studio to change it. With Pro, export clips as WAV, M4A (AAC) or AIFF files for podcasts, videos and audio editors.
Pronunciation (Pro)
When a name, brand or technical term comes out wrong, write it the way it sounds. "Nguyen" can become "win" and "SQL" can become "sequel". Vocalune uses your spelling every time, and you can listen before you save it.
Model Manager
Vocalune does not come with one fixed voice engine. You choose open AI text to speech models in the Model Manager, which shows the download size and memory each one needs before you download it. Every model uses a license that allows commercial use.
| Model | Languages | Download | Memory | Access | License |
|---|---|---|---|---|---|
| Kokoro 82M28 American and British English voices. Fast and light. | English | 342 MB | 8 GB | Free | Apache 2.0 |
| Pocket TTS23 voices. Each language is a separate download. | English, French, German, Spanish, Italian, Portuguese, Dutch | About 360 MB each | 8 GB | English free, other languages Pro | CC BY 4.0 |
| Qwen3-TTSNine voices. The 1.7B versions follow style directions such as "speak softly". | 10 languages | 2.0 to 3.1 GB | 8 GB, 16 GB for HQ | 0.6B free, 1.7B Pro | Apache 2.0 |
| Qwen3-TTS Voice DesignCreates a voice from a written description. | 10 languages | 2.3 or 3.1 GB | 8 GB, 16 GB for HQ | Pro | Apache 2.0 |
| Qwen3-TTS Voice CloningReads text in a voice from a short recording, with permission. | 10 languages | 2.0 to 3.1 GB | 8 GB, 16 GB for HQ | Pro | Apache 2.0 |
| VoxCPM2Describe a voice or clone one. Audio at 48 kHz. | 30 languages | 2.3 or 3.2 GB | 8 GB, 16 GB for 8-bit | Pro | Apache 2.0 |
Privacy
Speech is made by the model on your Mac. Vocalune connects to the internet only to download the models you choose and to talk to the App Store for purchases.
Pricing
There are no credits and no per-character fees. Prices below are in US dollars. The App Store shows your local price before you buy.
$0
$4.99 / month
or $29.99 / year, or $49.99 once for lifetime
Monthly includes a 3-day free trial for eligible new subscribers. Subscriptions renew automatically unless canceled at least 24 hours before the end of the current period. Manage or cancel them in your App Store account settings.
Requirements
FAQ
Vocalune is a text to speech app for Mac. You type, paste or import text, pick a voice, and it reads the text aloud with natural sounding AI voices. The voice models run on your Mac, not on a server.
Yes. You need an internet connection once to download a voice model. After that, speech is made on your Mac without a connection. The internet is only used again for new model downloads and App Store purchases.
No. Your text, your audio, your recordings and your settings stay on your Mac. Vocalune has no account, no analytics and no tracking.
Vocalune needs a Mac with Apple silicon (M1 or later) and macOS 26.2 or later. Every model has a version that runs with 8 GB of memory. The higher quality versions of some models need 16 GB.
There are 60 built-in voices across Kokoro, Pocket TTS and Qwen3-TTS, plus voices you design from a description or clone from a recording. Depending on the model, Vocalune speaks up to 30 languages.
Free includes Kokoro, Pocket TTS English, Qwen3-TTS 0.6B, up to 2,000 characters at a time, the Library and the Home dashboard. Pro adds texts of any length, audio export as WAV, M4A or AIFF, voice mixing, custom pronunciations, more languages, and the voice design and voice cloning models.
In the United States, Pro is $4.99 a month with a 3-day free trial for eligible new subscribers, $29.99 a year, or $49.99 once for lifetime access. Every plan unlocks the same features. The App Store shows the price in your local currency before you buy.
Only with their clear permission. Before Vocalune saves a cloned voice, you confirm that the recording is your own voice or that you have the speaker's permission. Cloned voices are stored on your Mac and are never uploaded.
Yes. Demo mode lets you try every screen right away. In Demo mode, Vocalune speaks with a voice built into macOS instead of the real model, so you can see how it works before you download anything.
Every model in Vocalune uses a license that allows commercial use (Apache 2.0 or CC BY 4.0). Some licenses ask for attribution. The model names and licenses are listed in Settings, About.
The App Store page is not live yet. Questions in the meantime? Write to [email protected].