Papercup’s award-winning machine learning engine produces synthetic voices that sound like human actors.
We’ve developed an award-winning machine learning text-to-speech system that has been backed by organizations like Innovate UK. Our in-house research team has published several papers, been granted patents and continues to be at the forefront of this new technology’s development.
The synthetic voices that our system produces are extremely lifelike and even capture some of the nuances of the original speaker’s vocal traits. The new voice is controlled and adapted by our translation team to make it indistinguishable from a native speaker of that language.
One of the key features of our patented speech synthesis solution is the range of voices and styles that we can generate. Our software gives you more control than ever before, meaning we can generate customized voices that suit each content creator or brand.
Tian is a senior ML engineer at Papercup and has been with us since 2019. In that time, she’s researched and deployed many iterations of our voice models, including one that can make Jesse say anything she wants.
Discover how our clients are using Papercup to unlock more value from their existing content.