Text to speech MP3. Make it, keep it, download it.

1. 1

   Write or paste the script

   Open the Studio and paste your text. Add pause and style tags if you want to direct the reading.

   

2. 2

   Choose a voice and generate

   Pick one of the 38 voices or one of your clones, then press Generate speech. A task starts, and you can keep working.

   

3. 3

   Play it and download

   The finished voice-over appears in your History. Press play to listen. Use the download button to get the MP3.

Each voice-over is its own task, so you can start the next one while the first is still finishing. That is handy when you have a list of scripts to turn into files.

You do not have to download a voice-over straight away. Everything you generate is kept in your History, in your browser, until you delete it. That is where you play, find and download your files.

Because History is stored in your browser, clearing your site data removes it. Download the files you care about.

Clonylab makes an MP3 first. When you choose WAV, the app converts that MP3 into a WAV file. So both files carry the same sound. The difference is the container.

More on the second one in text to WAV.

• Choose the voice first. Press play on a sample, then pick it.
• Write for the ear. Short sentences read better aloud than long ones.
• Add pauses. Put [pause 1s] where a speaker would breathe.
• Slow the important words. Wrap them in [slow] and [/slow].
• Listen before you share. Play the whole voice-over once at normal speed.

There is a full list of tags on the AI voiceover generator page.

On the free plan, credit “Made with Clonylab” when you share a voice-over. Pro and Team do not have that line. The standard voices are licensed for commercial use. Read text to speech for commercial use for the details, and the Voice licences page for each licence.

MP3s usually come in batches. A set of video intros, a series of lessons, a list of product lines. Here is a workflow that suits Clonylab’s task queue.

Batches need Pro or Team in practice, because the free plan has only 300 characters in total. See unlimited text to speech for what that covers.

More help: contact us with your browser and what you saw.

Once you make more than a handful of files, names and folders matter. The downloaded file name comes from your voice-over, so a little discipline before you generate saves time later.

Three things shape the sound, in this order.

• The voice. Different voices have different tone and pace, so choose first.
• The script. Short, clear sentences with natural pauses make the biggest difference.
• The settings. The sliders and tags fine-tune speed, pitch, pauses and volume.

Then listen once before you download. If a name is read oddly, respell it the way it sounds and generate again. You can read more about directing a reading on the AI voiceover generator page.

• Listen all the way through. Play the whole voice-over once, at normal volume.
• Check names and numbers. Respell anything that sounds wrong and generate again.
• Check the ends. The first and last words should not be cut or rushed.
• Match the volume. Use the Volume slider so the audio sits well with your music or video.
• Keep a copy. Download the file, and keep your script next to it.

These five checks take about a minute and catch almost every problem.

• Direct the reading with pauses and style tags.
• Choose a voice from the 38 English voices.
• Understand WAV files on the text to WAV page.

Questions about text to speech MP3