zerouploads

Text to Speech Converter

Turn text into spoken audio and download it as MP3 or WAV. No character cap, no watermark, no signup.

  • Preview
  • Unlimited
  • No signup
  • Private
  • Works offline

Text → Speech44 words · ≈ 00:17 at 1.0×

The text to read

3 sentencesplain text · numbers as words

Preview and voice

PreviewAmy · English (US) · 1.0×

Press Read aloud and the recording appears here, with a bar you can drag.

3 sentences readyWAV, then MP3

Voice42 available

  • Local
  • Local
  • Local
  • Local
  • Local
  • Local

6 of 42 listed18 languages

Voice settings

How fast the voice reads. It is set while recording, so press Read again after you change it.

Moves the voice up or down in semitones. Press Read again to hear a change.

Saved into the file, so the download is what you heard. Press Read again after a change.

Reading options

Amy is 60.18 MB, downloaded once and then kept on this device.

44 words · ≈ 00:17 at 1.0× · WAV 550 kB, MP3 275 kB

How it works

  1. 01

    Paste your text

    Any length. Paste a whole article, script or set of study notes.

  2. 02

    Pick a voice and a pace

    Local voices download once and then work offline. System voices need no download at all.

  3. 03

    Listen, then download

    Save an MP3 to share it. Save a WAV when the audio is going into an editor.

Local voices and system voices

The tool has two kinds of voice. Local voices are Piper models, 20 to 73 MB each, that download once and then run on your own processor. The first one also brings about 18 MB of pronunciation dictionaries. After that they work with no connection, and there is no per-character cost to pass on.

System voices come from your operating system, through the Web Speech API. They need no download and start straight away. But they read through your speakers and give back no file, so there is nothing to save. How good they sound depends on the voices your system has.

Neither kind sends your text anywhere. The list only shows system voices that run on your device, so a browser's online voices are left out. That makes the tool safe for confidential text, and it is why there is no length limit.

Length
No character limit. The tool reads long text in pieces of up to 300 characters, one after another.
Voices
42 local voices across 18 languages and regions, plus the system voices that run on your device.
Output
WAV, or MP3 at 128 kbps. No watermark.
Offline
Once a voice is cached, reading works with no connection at all.

Getting natural-sounding output

Punctuation does most of the work. Commas and full stops become pauses. Text written to be read often runs long clauses together, and it sounds far better once you split them into separate sentences.

Read numbers as words is on by default, so 1999 is read as nineteen ninety-nine. That is right for prose and wrong for a serial number. Turn it off and every run of two or more digits is read one digit at a time.

Acronyms are the other common snag. The voice says some as words and spells out others. Spell out acronyms puts full stops between the letters of any word in two to six capitals. Then it is always read letter by letter.

Frequently asked questions

Is my text sent to a server to be spoken?
No. Local voices run on your own processor. The only system voices listed are ones your device runs itself, so your text never leaves the machine.
Is there a character limit?
None. The work happens in this tab, so there is no per-character cost for anyone to cap.
Can I use the audio commercially?
It depends on the voice. Each Piper voice was trained on a different recording set with its own licence, and some licences limit commercial use. Check the voice's model card in rhasspy/piper-voices on Hugging Face first. For system voices, check the terms of your operating system.
Why does a voice need to download?
A local voice is a model of 20 to 73 MB. It downloads once and your browser keeps it, so the next use starts straight away and needs no connection.
Which voices work offline?
Every local voice you have already used, and every system voice in the list. The list leaves out system voices that need the network.
Can I control pronunciation?
Often, yes. Spelling a word the way it sounds usually fixes it. Spell out acronyms puts full stops between capitals, so they are read as letters.

More audio tools

Age CalculatorWork out exact age in years, months and days between any two dates, plus totals and the next birthday.AVIF to JPGConvert AVIF images to JPG in your browser. AVIF packs tighter than JPG, so expect a bigger file, and clear parts take the colour you pick.Base64 DecodeDecode Base64 back to text or a file. Handles URL-safe input, missing padding and embedded line breaks.Base64 EncodeEncode text or files to Base64 in your browser. Standard or URL-safe alphabet, with or without padding.BMI CalculatorWork out body mass index from height and weight, in metric or imperial, with the arithmetic shown.Body Fat CalculatorEstimate body fat percentage from tape measurements, with lean mass, fat mass and the full working shown.
Browse all audio tools