paulowoicho/t5-podcast-summarisation

The model is based on Google's T5 which was pretrained on the C4 dataset . The model can be used with the summarisation as follows: . It can also learn to generate promotional material (links, hashtags, etc) in its summaries, as such as links and hashtags . If using on Colab, the instance will crash if the number of tokens in the transcript exceeds 7000. The model generated reasonable summaries even when the podcast transcript was truncated to reduce the size of the tokens. It also generated reasonable summararies when the transcript was reduced to less than 5,000 tokens per hour of the original length of the transcript. If using Colab the instance would,