Introduction

See https://github.com/k2-fsa/sherpa

This repo uses onnxruntime and does not depend on libtorch.

We provide exported models in onnx format and they can be downloaded using the following links:

• English: https://huggingface.co/csukuangfj/icefall-asr-librispeech-pruned-transducer-stateless3-2022-05-13
• Chinese: TODO

NOTE: We provide only non-streaming models at present.

HINT: The script for exporting the English model can be found at https://github.com/k2-fsa/icefall/blob/master/egs/librispeech/ASR/pruned_transducer_stateless3/export.py

Usage

git clone https://github.com/k2-fsa/sherpa-onnx
cd sherpa-onnx
mkdir build
cd build
cmake ..
make -j6
cd ..

Download the pretrained model (English)

Caution: You have to run git lfs install. Otherwise, you will be SAD later.

git lfs install
git clone https://huggingface.co/csukuangfj/icefall-asr-librispeech-pruned-transducer-stateless3-2022-05-13

./build/bin/sherpa-onnx --help

./build/bin/sherpa-onnx \
  ./icefall-asr-librispeech-pruned-transducer-stateless3-2022-05-13/data/lang_bpe_500/tokens.txt \
  ./icefall-asr-librispeech-pruned-transducer-stateless3-2022-05-13/exp/onnx/encoder.onnx \
  ./icefall-asr-librispeech-pruned-transducer-stateless3-2022-05-13/exp/onnx/decoder.onnx \
  ./icefall-asr-librispeech-pruned-transducer-stateless3-2022-05-13/exp/onnx/joiner.onnx \
  ./icefall-asr-librispeech-pruned-transducer-stateless3-2022-05-13/exp/onnx/joiner_encoder_proj.onnx \
  ./icefall-asr-librispeech-pruned-transducer-stateless3-2022-05-13/exp/onnx/joiner_decoder_proj.onnx \
  ./icefall-asr-librispeech-pruned-transducer-stateless3-2022-05-13/test_wavs/1089-134686-0001.wav