Create TTS Voice

1. Login or Create An Account
2. Create TTS Voice Clone
3. Upload lots of sounds (or create many recordings with your microphone) with clear speech, character-like lines, and plenty of phrases.
4. Click the "Create Voice Clone" button on the soundboard page
5. Enjoy your new TTS voice!

Trim clips

Trim the clips to only contain speech. Use the button, or audio editing software like Audacity.

Multiple clips are better than one short sample

You should upload lots of clips of clear speech rather than one short sample. Aim for as much clear speech as possible across many clips, with lots of recognizable words and a wide variety of phrases and lines. Trim your clips to your best, clearest speech and use our free audio editor or software like Audacity.

Got lots of samples? Make multiple voices

If you have many clips of different expressions, consider splitting them into separate soundboards and cloning each one - e.g. a Normal voice, an Angry voice, a Surprised voice. Each will be a distinct, more complete TTS voice with lots of usable content.

Words must be recognizable

The voice cloning process recognizes the words spoken in your clips with speech-to-text. If they cannot be understood (e.g. mumbles, screaming), your voice will not be optimal.

Use different intonations and emotions

Use clips with different intonations and emotions to give the AI further information about your voice. If you want your voice to have a specific intonation or emotion, use clips consistently reflecting that style.

Use different words and sentences

Use clips with with a variety of words and sentences.

Remove background music or sound effects

Make sure the clips contain only speech, no background music or sound effects. MVSEP is a useful tool to separate voices from music.

Use a good microphone

Use a good microphone to record your voice. The better the quality of the recording, the better the results.

Original sounds will be deleted

Your original sounds will be deleted when cloning. Keep a backup if they are important.

Crazy voices

If you want to make crazy nonsense voices, the source audio still needs plenty of recognizable spoken words across multiple clips for cloning to work well.

Use a clear name for your voice or board

Choose a descriptive name for your voice or board so the AI can better understand the style, character, or tone you want it to learn from your samples and build a fuller voice.

Keep trying

Don't be discouraged if the first attempt doesn't sound perfect. Try again with different clips and settings. Have fun!

Example sentences

Here are some example sentences you can use to create your TTS voice. You only need to say one. Don't forget to speak in the intonation and emotion you want in your TTS voice, or it may come out monotone. These are called the 'Harvard Sentences'.

• The birch canoe slid on the smooth planks.
• Glue the sheet to the dark blue background.
• It's easy to tell the depth of a well.
• These days a chicken leg is a rare dish.
• Rice is often served in round bowls.
• The juice of lemons makes fine punch.
• The box was thrown beside the parked truck.
• The hogs were fed chopped corn and garbage.
• Four hours of steady work faced us.
• A large size in stockings is hard to sell.

Have fun!