In just a few short months, all kinds of generative AI possibilities have emerged from text prompts. Whether it is music with lyrics, new video tools, or music videos that combine the two, generative AI is rapidly expanding.

Still, it can be difficult to achieve dialogue (outside of more robotic text-to-speech options and dedicated chatbots); ElevenLabs has one of the best synthetic voice models available, including clones of real voices (permission and licensing agreements required). Its latest project goes a step further and allows you to design your own voice from scratch.

While I believe that lending voices to characters should remain the specialty of trained voice actors, it is not always feasible, and ElevenLabs' new voice design engine can create voices from text prompts in a matter of seconds and and can voice everything you want it to.

This is useful when a project requires a very specific voice, or when you need to change what is said in real time as the game flows.

If you have played Dungeons & Dragons, you know that the right voice can make all the difference when setting up a scene.

Voice design tools not only help players create their own deep backstories and lore, but also democratize the technology for indie developers who are working alone on ambitious new projects.

When coupled with AI video generation tools such as Eleven Labs' own, it could allow film students to create worlds, characters, and interactions all by themselves within the same project.

The Eleven Lab example shows that more detailed prompts are better. Professional, relaxed, assertive. Stereotypes abound, as no one in the UK will call you “Old Bean” anymore.