gianpaolocoro/nvidia_stt_it_fastconformer_hybrid_large_pc repository overview

From the folder where the audio.wav is, run:

docker run --runtime=nvidia --gpus all -v ${PWD}:/home/docker -it gianpaolocoro/nvidia_stt_it_fastconformer_hybrid_large_pc:1.2 /bin/bash -c "export LD_LIBRARY_PATH=/usr/local/cuda-11.8/targets/x86_64-linux/lib:$LD_LIBRARY_PATH && python doasr_home.py /home/docker/audio.wav"

This is based on pytorch and is an all-in-one version. It was developed on a Linux Ubuntu 22.04 version endowed with a TitanXP graphic card.