The audio tracks are usually handled with Adobe Enhance, which works in the browser as part of Adobe subscriptions (which are still, I emphasize, still active :D).
@Tomi_Valkeajarvi recently cleverly coaxed Claude into making podcast cuts based on the audio tracks by giving the language model access to the files, and that initial work could then be fine-tuned in Premiere Pro. He can tell you more about it! 
As far as I understand, these are actual professional tools and we are on the cutting edge here… 
The community team has Macs, so M-series chips are our power, not NVIDIA.