Text to Speech
Type or paste text, choose a voice, and download natural-sounding speech as MP3 or WAV. The neural voice runs entirely on your device, so your text is never uploaded.
Reviewed by Ismail Sab Ajnalkar · Updated July 2026
File Transmute's text-to-speech uses Kokoro, an 82-million-parameter neural voice model that runs in your browser through WebAssembly. Unlike cloud TTS services that send your text to a server (and meter every character), File Transmute synthesizes speech locally — the model downloads once on first use, then works privately and offline. Choose from American and British voices and export studio-quality MP3 or WAV.
Best for
- Listening to an article, script, or notes instead of reading them
- Creating a voiceover or audio version of written content
- Proofreading by ear — hearing text read back catches errors
Good to know
Turns text into spoken audio on your device using browser and on-device voices — nothing is uploaded. Voice range and naturalness depend on the voices available in your browser and operating system. Very long text uses more memory, so break book-length input into sections.
How to convert Text → Speech
Type or paste your text — or upload a .txt file.
Pick a voice and choose MP3 or WAV output.
Click Convert; the audio is generated on your device and downloads instantly.