Summarization of long Transcripts

[NightmareQueenJune]

I am trying to summarize very long (8-12 hour long) transcripts (the one I am testing is around 8 hours and 36000 words long). This of course does not fit into the context length of the models I am loading via ollama. How do these transcripts get loaded into ollama?
As far as I can see the front of the document is truncated. Would there be the possibility to somehow implement a multi-step summarization, where the transcript is cut into pieces fitting the context length of the model, each getting summarized?