This is the runtime of WeNet.
We are going to support the following platforms:
- Various deep learning inference engines, such as LibTorch, ONNX, OpenVINO, TVM, and so on.
- Various OS, such as android, iOS, Harmony, and so on.
- Various AI chips, such as GPU, Horzion BPU, and so on.
- Various hardware platforms, such as Raspberry Pi.
- Various language binding, such as python and go.
Feel free to volunteer yourself if you are interested in trying out some items(they do not have to be on the list).
For each platform, we will create a subdirectory in runtime. Currently, we have:
- LibTorch: in c++, the default engine of WeNet.
- OnnxRuntime: in c++, the official runtime for onnx model.
- GPU: in python, powered by triton.
- android: in java, it shows an APP demo.
- Language binding
- binding/python: python is the first class for binding.
- binding/go: ongoing.
Please note that server and device are depreated, and they will be removed in the future.