An updated version of the “ElevenLabs” neural network has been unveiled. It can now generate any sound, music, and speech.
In particular, the neural network generates birdsong, rain, and sounds of various processes based on a single text description.
This neural network can convert any voice into another voice in 29 languages. In addition, after conversion, all emotions and the voice’s timbre are preserved, and the voice quality is further improved. The converted voice is almost indistinguishable from real speech.
Prepared by Bahora Saidova.
