Starting from version v3.85, real-time speech-to-text has been added, and the noise reduction model and voice separation model have been updated. To avoid making the software package too large, the required models are not included by default. You need to manually download and extract them, then place the extracted files into the models/onnx folder within the software directory.
GitHub Download Address / Download from github.com
https://github.com/jianchang512/stt/releases/download/0.0/noise-uvr-speaker-realtime.7z