OpenVoice V2 is a text-to-speech (TTS) model released in April 2024, which includes all the features of V1 and has been improved. It employs a distinct training strategy to deliver superior sound quality, supporting English, Spanish, French, Chinese, Japanese, and Korean, among other languages. Additionally, it provides free usage for commercial purposes. OpenVoice V2 can precisely clone reference pitch coloration and generate speech in various languages and accents. It also supports zero-shot cross-language cloning, meaning the language of the generated speech and the reference speech do not need to be present in a large-scale multilingual training dataset.