Very nice indeed. Thanks for sharing. You mention that you use whisper on the cloud. I was wondering what you think of on-device decoding using the base.en mod

Very nice indeed. Thanks for sharing. You mention that you use whisper on the cloud. I was wondering what you think of on-device decoding using the base.en model and tflite. Something along the lines of: https://colab.research.google.com/github/usefulsensors/openai-whisper/blob/main/tinynn_pytorch_to_tflite_int8.ipynb#scrollTo=DeDY0j0zitxx https://colab.research.google.com/github/usefulsensors/opena...