run-whisper.sh
520 字节
#!/usr/bin/env bash
# Please refer to
# https://k2-fsa.github.io/sherpa/onnx/pretrained_models/whisper/tiny.en.html
# to download the model
# before you run this script.
#
# You can switch to a different offline model if you need
./non-streaming-decode-files \
--whisper-encoder=./sherpa-onnx-whisper-tiny.en/tiny.en-encoder.onnx \
--whisper-decoder=./sherpa-onnx-whisper-tiny.en/tiny.en-decoder.onnx \
--tokens=./sherpa-onnx-whisper-tiny.en/tiny.en-tokens.txt \
./sherpa-onnx-whisper-tiny.en/test_wavs/0.wav