实时语音模型

Real-time speech models can recognize and process speech as the user speaks. It is mainly used in applications such as speech-to-text, voice assistants, and real-time translation. By utilizing deep learning algorithms, the model can quickly and accurately convert speech signals into text or other forms of output, providing an efficient user interaction experience.

The following models are now available for purchase:

• MaaS-4o-realtime-preview