Automatic Speech Recognition (ASR) with Facebook AI's wav2vec 2.0 model using Huggingface
Facebook AI's wav2vec 2.0 is a new framework that claims to perform Automatic Speech Recognition without using a language model. In this video we will quickly take a look at the abstract of the paper and then move on to the implementation of this system using Huggingface. Huggingface provides us with wav2vec2-base-960h model that can be used to perform ASR. As described in the video, here are the relevant links:
1. Link to the paper - https://arxiv.org/abs/2006.11477
2. Link to Huggingface's wave2vec 2.0 model page - https://huggingface.co/facebook/wav2v...
3. Link to the Colab notebook - https://colab.research.google.com/dri...