MyShell releases open source AI voice cloning tool OpenVoice, targeting the field of voice imitation

Canadian AI startup MyShell, in collaboration with Massachusetts Institute of Technology (MIT) and Tsinghua University, has launched OpenVoice, an open-source AI voice cloning tool that has garnered a lot of attention in the field of voice mimicry.OpenVoice employs a highly efficient methodology to rapidly clone a user's voice and significantly reduce the consumption of computing resources. Users can generate personalized voice clones with a simple operation that does not require reading specific text aloud, but simply speaking randomly for a few seconds. The core technology behind the tool includes text-to-speech (TTS) models and pitch shifter models, which are trained with a large number of audio samples, making OpenVoice far less computationally expensive than other methods for cloning speech.