Machine Learning is Fun Part 6: How to do Speech Recognition with Deep Learning
Adam Geitgey

Such great writing here and a brilliant piece of work! Adam, I was wondering if one would use the Fourier transform if the recognition was on fixed length sounds instead of speech? It seems you could preprocess the entire sound data as the entire input vector… provided the sound wasn’t very long of course?

One clap, two clap, three clap, forty?

By clapping more or less, you can signal to us which stories really stand out.