models analyze vocal files.
Remove buzzing, chatter, and white noise.
both noisy and clean vocal files
The motivation behind The Clarity Project is to create clear and concise audio from a lecture.
There are many class lectures that are recorded for re-listening at a future date. The problem is that these recordings are usually done in a single take and are frequently riddled with background noise and distractions. Buzzing, chatter, white noise, and other unwanted sound make it difficult to focus on the speaker, which is where The Clarity Project comes in.
The question is: How can lectures be made into clear, more concise audio outside of traditional methods such as high or low pass filters? Additionally, can traditional methods be bested by a newer method?
400
150k
The model uses sequence to sequence algorithms to try and find a mathematical way to manipulate the array of waveforms from the noisy audio
Let Us Know if you have any questions