Acknowledge.md

Acknowledge

1. We borrowed a lot of code from Kaldi for data preparation.
2. We borrowed a lot of code from ESPnet. FunASR follows up the training and finetuning pipelines of ESPnet.
3. We referred Wenet for building dataloader for large scale data training.
4. We acknowledge ChinaTelecom for contributing the VAD runtime.
5. We acknowledge RapidAI for contributing the Paraformer and CT_Transformer-punc runtime.
6. We acknowledge AiHealthx for contributing the websocket service and html5.
7. We acknowledge XVERSE for contributing the grpc service.