MARS5 an opensource TTS model to replicate performances (from 2-3s of audio reference) in 140+ languages, even for extremely tough prosodic scenarios like sports commentary, movies, anime & more. Join our Discord https://discord.gg/4GVdQ28cZC today!
CAMB.AI introduces MARS5, a fully open-source (commercially usable) TTS with break-through prosody and realism available on our Github: https://www.github.com/camb-ai/m...
Why is it different?
MARS5 is able to replicate performances (from 2-3s of audio reference) in 140+ languages, even for extremely tough prosodic scenarios like sports commentary, movies, anime and more; hard prosody that most closed-source and open-source TTS models struggle with today.
We're excited for you to try, build on and use MARS5 for research and creative applications. Let us know any feedback on our Discord!
Highlights:
Training data: Trained on over 150K+ hours of data.
Params: 1.2 Bn (750/450)
Multilingual: Open-sourcing in English to begin with, but can access it in 140+ languages on camb.ai
Diversity in prosody: can handle very hard prosodic elements like commentary, shouting, anime etc.
Awesome launch Akshat! Congratulations. Do you also provide a hosted option and an API offering?
Well done on the impressive launch! The large training dataset is evident in the model's accuracy. I'm glad to see that it supports various languages.
Congratulations on the launch! Have you conducted a benchmark comparison between Mars5 and one of those the market leader, like Elevenlabs, for their prosodic TTS services? I'm eager to see a comprehensive analysis!
Congrats on the launch! How does this application handle speaker consistency and identity preservation?
Cool stuff! Congrats on the launch guys! Always nice to see new open source products launching .
Love the innovative TTS model! Excited to see its impact on diverse audio projects. Keep it up!
