#Live ASR Engine

I have implemented (not from scratch) LiveASREngine using whisper using the following codebase:

https://github.com/oliverguhr/wav2vec2-live

Installation requirements for whisper (developing anaconda environment):

1. conda create --name tips -c nvidia -c pytorch -c huggingface -c defaults -c conda-forge python autopep8 pytorch torchvision torchaudio
2. pytorch-cuda=11.7 transformers pyaudio webrtcvad rx halo onnx onnxruntime pyctcdecode
3. pip install pysoundfile
4. conda install -c conda-forge ffmpeg