AI Voice
An AI voice generator that runs on your PC
Murmur turns scripts into voiceovers with local models. Pick a built-in voice, clone a reference, or describe a voice, then export WAV.
Free plan with Kokoro voices, scripts up to 1,500 characters · 3-day Pro trial, then $49 once
AI voice generator Windows
Murmur runs locally on 64-bit Windows 10 and 11, so your text stays on your PC and you can export WAV without cloud APIs or per-word fees.
- Local AI voice generation
- Voice cloning and Voice Design
- Kokoro included offline
- One-time $49
01 · Context
What an AI voice generator does
An AI voice generator turns written text into spoken audio. Modern models can sound natural, follow punctuation, and handle long scripts. Some can also clone a voice from a sample or create a new voice from a description.
Cloud generators run the model on a server and charge by subscription or usage. Murmur runs the model on your Windows PC and charges once. That changes the workflow: no character meter, no upload step for generation, and a file you keep.
02 · Highlights
What you can make on Windows
YouTube and social voiceovers
Generate narration for videos, shorts, and explainers, then export WAV for your editor.
Course and training narration
Turn lesson scripts into audio and regenerate a line when the copy changes.
Podcast intros and ads
Audition a few voices, pick one, and export a clean read.
Accessible versions of written content
Create listenable versions of articles, guides, and internal documents.
Character and brand voices
Use Voice Design or a cloned reference to keep a consistent voice across a series.
03 · Details
Built-in voices, cloning, or Voice Design
| Approach | What you need | Windows models |
|---|---|---|
| Built-in voices | Nothing extra | Kokoro, with 54 voices in nine languages |
| Voice cloning | A clean reference clip | Chatterbox, Qwen3-TTS Base, SparkTTS, Breeze TTS 2, OmniVoice |
| Voice Design | A written description | Qwen3-TTS Voice Design, Breeze TTS 2 |
04 · Workflow
A script-to-export workflow
05 · Context
What the Windows beta does not do yet
The Windows beta is a single-script composer. It does not have projects, a batch queue, community voice browsing, CLI or MCP automation, MDM deployment, M4A export, or automatic updates. Fish S2 Pro is marked Coming soon on Windows.
The Mac app adds those production features. If you need them, use the Mac app. If you need a local Windows voice generator for one script at a time, the beta covers the core job.
FAQ
Common questions
Yes. Murmur generates speech from text with local AI models. It can use built-in voices, clone a reference clip, or design a voice from a description with a compatible model.
Windows · Local TTS
Make your next voiceover on Windows
Download Murmur, generate with Kokoro offline, and try the Pro models free for 3 days. Pro is $49 once.