Chapter
The Future of Deep Learning and Audio
56:27 - 1:01:27 (04:59)
This episode discusses the potential for cheaper lenses combined with intentional movement and how deep learning is evolving to allow for better subpic and resolution. The guest also shares his findings on super convergence and the possibility of automatically combining audio from multiple sources to remove noise in devices like Alexa.
Clips
The focus should be on generating high-quality audio and video using less expensive equipment like cheap microphones and a bit of intentional movement, which provides enough information for excellent subpics or deep learning.
56:27 - 59:42 (03:14)
Summary
The focus should be on generating high-quality audio and video using less expensive equipment like cheap microphones and a bit of intentional movement, which provides enough information for excellent subpics or deep learning. The same approach used for super resolution in generating high-quality images could also work for audio.
ChapterThe Future of Deep Learning and Audio
EpisodeJeremy Howard: fast.ai Deep Learning Courses and Research
PodcastLex Fridman Podcast
The discovery of super-convergence in neural networks allows for the training of networks 10 times faster than before.
59:42 - 1:01:27 (01:44)
Summary
The discovery of super-convergence in neural networks allows for the training of networks 10 times faster than before. This phenomenon has been observed through the use of certain high parameter settings.