| Microsoft Unveils Open-Sourced VALLE-X for Voice Cloning Microsoft has released a new open-sourced text-to-speech synthesis and voice cloning model, VALLE-X. This new model has been designed to help companies develop multilingual speaking applications. It can learn to clone a person's voice and generate human-like speech in multiple languages. The release of VALLE-X is part of Microsoft's broader effort to create a versatile and powerful AI solution for text-to-speech synthesis. VALLE-X is based on the state-of-the-art Deep Voice 3 architecture, which uses deep neural networks to generate synthetic speech. The model is capable of generating speech in multiple languages and supports a wide range of voice styles. The open-sourced model is expected to make it easier for developers to create apps that use voice cloning to transform text into human-like sounding speech. This could open the door to a wide range of potential uses, including developing more natural sounding customer service bots, creating interactive intelligent virtual assistants, and improving accessibility through speech synthesis. SinceVALLE-X is open-sourced, developers have access to the code and data for the model. This will enable them to extend and customize the model to create unique applications. Microsoft is hoping that this will spur innovation and allow developers to create more advanced voice cloning solutions. |