Free online tools

🌐 English ▾

☰

Piper + Kokoro · local processing

AI text to speech

Choose an engine and a voice, generate AI speech in your browser, and download it as WAV.

Text76 / 5,000characters

The long texts are automatically divided into fragments. Piper is the recommended option for Spanish; Kokoro offers more variety of English voices.

Privacy:the text is processed locally. Only the necessary files of the engine and the voice that you select are downloaded. Your text is not sent to UtiliTools Pro servers.

Convert text to speech directly in your browser

The tool allows you to choose between Piper and Kokoro. Piper offers different voices in Spanish, including two options from Mexico, while Kokoro adds a wider collection of English voices. The models run directly in your browser.

Two AI engines

Choose between Piper and Kokoro according to language and voice style.

Latin American Spanish

Piper includes two voices from Mexico in addition to several options from Spain.

More English voices

Kokoro offers several American and British voices.

Local privacy

The text does not need to be sent to UtiliTools Pro servers.

Long texts

The content is automatically split into chunks for processing.

Download WAV

Play the result and save it as WAV audio.

How to convert text to speech

1
Write or paste the text you want to convert to audio.
2
Choose the Piper or Kokoro engine and then select a voice.
3
Click “Generate speech with AI” and wait for processing to finish.
4
Listen to the result and download the WAV file.

No paid API

Piper and Kokoro can run locally using web technologies such as WebAssembly and ONNX. Only the files needed for the selected engine and voice are downloaded. There is no charge per character or generation.

The first download is the largest

Switching engines or choosing a new Piper voice may require an additional download. Kokoro voices share a common model. Files cached by your browser can speed up future generations.

Frequently asked questions

Is the text-to-speech tool free?

Yes. Speech synthesis runs in your browser with Piper or Kokoro, without an API that charges for each generated audio file.

Is my text sent to a server?

It is not sent to UtiliTools Pro servers. Your browser downloads the selected voice model and generates the audio locally.

What languages are available?

Piper includes voices from Mexico, Spain, and the United States. Kokoro adds several American and British voices. Kokoro is used here for English because its JavaScript wrapper does not yet reliably convert Spanish text to phonemes.

Why does it take longer the first time?

Piper downloads the selected voice, while Kokoro downloads its shared model and the required voice data. The first generation may take longer; afterward, your browser can reuse cached files.

Can I download the audio?

Yes. After generating the voice you can play it on the page and download it as WAV file.

Can I convert long texts?

Yes. The tool automatically divides the text into fragments to process them one by one and then combines them into a single audio file.