Are you launching an AI Spotify competitor and want API access to a community-curated ever-improving list of voice models without having to set up serverless GPUs? Are you a producer who wants to layer guide vocals on your tracks before sending them to a vocalist? Do you want to just make fun memes for your friends? Well look no further. Uberduck began hosting text-to-speech models back in the AI Juice Wrld era (2021). Along the way, we’ve learned about everything from training large zero shot models to personalizing lyrics for marketing, and maintained our commitment to hosting contributed models in production on a uniquely large scale. Over the last couple of months, new voice-to-voice architectures like soViTS-svc and RVC have amazed the music community, and we’ve added support for submission of these models to our site, receiving hundreds of submissions. We are able to offer many voices with minimal warmup time, and our list of voices and model types is always improving due to the community-managed nature of our product. We also have a proprietary rap-generation API that we think is pretty quacking good. Thanks for stopping by.

Congratulations on the launch! This is an Uber (forgive the pun) cool product 🔥🔥🔥. Haven't seen anything like this so far. I have some fun use cases in mind. Questions: - Are you currently able to train/generate rap lyrics on someone's voice (e.g. mine?). - What are you shipping next? I'll certainly try out the product soon. Well done again!

Uberduck is the best duck in the pond. It's not only great for super affordable TTS but they also have all kinds of other api's like rap and voice to voice, it's really refreshing! Being a bird myself, I'm happy to call uberduck one of my biggest buddies, quack quack!

This is so cool! Congrats on the launch :D

I've had a lot of fun generating voice prompts with Uberduck! Very seamless and easy to use.