AI text-to-speech is getting scary good...

Introduction

AI voice generators have always been a topic of avid discussion. Yet, they often come with hefty price tags or subpar quality. Fortunately, that narrative is changing. Enter Parlor Text-to-Speech, the ground-breaking open-source model designed to sound like a genuine audiobook narrator.

It's important to note that this is only the first iteration of Parlor's model. To provide an example, I selected one of the available presets and added custom text. Additionally, you can use the description box to input prompts, allowing the model to generate text in different voices.

After hitting generate, the output is quite impressive.

"Remember, this is only the first iteration of the model, so maybe we will hear more about them on the news you can use on the Adantage channel."

There were some issues with proper name recognition. For example, it struggled with the word "YouTube." Overall, the quality was incredible, considering it's still an early version. You can try it out yourself and experience the future of text-to-speech technology.