Repository files navigation

sherpa is an open-source speech-text-text inference framework usingPyTorch, focusingexclusively on end-to-end (E2E) models,namely transducer- and CTC-based models. It provides both C++ and Python APIs.

This project focuses on deployment, i.e., using pre-trained models totranscribe speech. If you are interested in how to train or fine-tune your ownmodels, please refer toicefall.

We also have othersimilar projects that don't depend on PyTorch:

• sherpa-onnx
• sherpa-ncnn

sherpa-onnx andsherpa-ncnn also support iOS, Android and embedded systems.

Please refer to thedocumentation athttps://k2-fsa.github.io/sherpa/

Trysherpa from within your browser without installing anything:https://huggingface.co/spaces/k2-fsa/automatic-speech-recognition