How to get a soft and cute female voice with OpenAI TTS?

Hi,

I’m working on a project to help users learn Japanese, and I’m using the gpt-4o-mini-tts model to generate spoken responses.

I would like the voice to be female, soft, and cute, with a friendly and encouraging tone, but I’m not sure how to configure it properly.

My questions:

  1. Which built-in voice is best suited for a soft and cute female style?

  2. What kind of instructions in the instructions field can make the voice more warm and charming?

  3. Are there tips for adjusting pitch, speed, or tone to achieve this effect?

Thank you in advance for your advice!

Hi,

Welcome to the forum.

You could check this page https://www.openai.fm/

1 Like

Welcome to the community.

There are some examples on openai.fm that you can use to develop your own style.

In the github repository, you can directly find a list of all instructions and styles available to help give you some ideas.

1 Like