A Fundamental End-to-End Speech Recognition Toolkit and Open Source SOTA Pretrained Models, Supporting Speech Recognition, Voice Activity Detection, Text Post-processing etc.
The parameter "quantize" is not used when starting the server. For example, if I set quantize=false, the runtime_sdk_download_tool.py will still search model_quantize.onnx. This PR is trying to refactor "--quantize" to each branch in the code.
The parameter "quantize" is not used when starting the server. For example, if I set quantize=false, the runtime_sdk_download_tool.py will still search model_quantize.onnx. This PR is trying to refactor "--quantize" to each branch in the code.