Type or paste text, then press Read aloud. Nothing is uploaded.
- Speech made on your deviceOpen-source voices in your browser
- 0 uploadsYour text never reaches a server
- Download WAVUse it in videos or slides
- No sign-upNo account needed
How to do it
Paste your script, choose a voice that suits your channel, and generate. Listen through once, fix any mispronounced words by spelling them the way they sound, then download the WAV and drop it into your editor under your footage.
How it works
- Type, paste or open a .txt or .pdf file (text PDFs only).
- Choose one of 13 English voices and a speed.
- Press Read aloud. The first time, the voice model downloads once (about 90 MB) and is cached.
- Speech is generated on your device part by part and starts playing straight away. When it finishes, download the WAV file.
Tips for natural-sounding speech
| Problem | What to do |
|---|---|
| A name is mispronounced | Spell it the way it sounds, e.g. "Siobhan" as "Shiv-awn" |
| No pause where you want one | Add a full stop, or a blank line for a longer pause |
| Numbers read oddly | Write them in words: "twenty twenty-six", "three point five percent" |
| Abbreviations read letter by letter | Write the full words: "for example" instead of "e.g." |
| Sounds rushed | Lower the speed to 0.9 and use shorter sentences |
Honest limits
- English only, with American and British voices.
- Speed depends on your device. On a typical laptop, expect very roughly real-time speed or faster; phones are slower.
- The voices do not do emotions on command or sing.
- Output is WAV (about 2.9 MB per minute). Convert to MP3 in any audio editor if you need a smaller file.
Your text stays private
Most free text-to-speech sites send your text to a server. VoiceRuler generates the voice inside your browser, so your words never leave your device. The Kokoro model is released under the Apache-2.0 licence, which allows commercial use of the output.
Related guides
- Text to speech for TikTok and Reels narration
- Free AI voice-over generator
- Text to speech for audiobooks
- Listen to an ebook with text to speech
- Narrator voice text to speech
- Text to speech for presentations and slides
- Text to speech for a podcast intro
Frequently asked questions
Can I monetise videos that use these voices?
The Kokoro model is released under the Apache-2.0 licence, which allows commercial use. YouTube's rules still apply: channels need original, valuable content, and YouTube asks creators to disclose realistic synthetic content in some cases. Read YouTube's current policies for your channel.
Is there a character limit?
You can process up to 100,000 characters at a time. Long scripts are generated in parts and joined into one file.
Is VoiceRuler free?
Yes. There is no sign-up, no daily limit and no watermark, because the voice runs on your device.
Is my text uploaded?
No. Speech is generated in your browser. Only the model and voice files are downloaded.
Which languages are supported?
English only, in American and British voices.