"TF-TRT Warning: Could not find TensorRT" in Google Colab

[7k50]

Quick remark (I'm not sure if this is pertinent, as I don't know if WhisperX uses Tensorflow in any meaningful way?):

I get the below warning while running WhisperX in Google Colab:

W tensorflow/compiler/tf2tensorrt/utils/py_utils.cc:38] TF-TRT Warning: Could not find TensorRT

How I install WhisperX in Google Colab currently:

!pip install git+https://github.com/m-bain/whisperx.git
!pip install light-the-torch

#conda install pytorch==2.0.0 torchvision==0.15.0 torchaudio==2.0.1 pytorch-cuda=11.8 -c pytorch -c nvidia
!ltt install --pytorch-computation-backend=cu118 torch==2.0.0 torchvision==0.15.0 torchaudio==2.0.1

I run WhisperX via:

!whisperx "{audio_path_wav}" \
    --task "transcribe" \
    --model "{model_name}" \
    --language "{language}" \
    --output_dir "{dir_whisper}" \
    --device "cuda" \
    --align_model "WAV2VEC2_ASR_LARGE_LV60K_960H" \
    --diarize \
    --min_speakers "{num_speakers}" \
    --max_speakers "{num_speakers}" \
    --output_format "srt" \
    --highlight_words True \
    --hf_token "{hf_token}"

[sorgfresser]

WhisperX uses Pytorch, as such this shouldn't be a problem. Don't think it affects performance etc. in any way.