One of the drawbacks to streaming television and movies is at times, buffering issues cause out-of-syncing between audio and video, stutters, stalls, or pixilation. An increased amount of this interference tends to cause viewers to tune out of what they are attempting to stream.
Professor at the Massachusetts Institute of Technology (MIT), Mohammad Alizadeh explained,
“Studies show that users abandon video sessions if the quality is too low, leading to major losses in ad revenue for content providers. Sites constantly have to be looking for new ways to innovate.”
According to MIT’s Computer Science and Artificial Intelligence Laboratory (CSAIL), AI is the answer. Streaming services typically use ABR (Adaptive BitRate) algorithms when determining the current resolution for video playback on the system of the viewer. As we stream something, a database chops the data we see into smaller bits to send us what we need in order to receive a clear picture for the immediate viewing experience.
As the algorithms are being transmitted to your device, they create a buffer ahead of the other so it doesn’t create an overal buffering issue that results in a stall. When figuring out the buffering ratio, the algorithm tends to miss sudden changes in network conditions, which ends up in pixilation and loss of picture quality.
To combat this problem, CSAIL has created an artificial intelligence neural network called, Pensieve.” This form of intelligence will pick algorithms based on ever-changing network conditions. With the use of AI, the switch between rate-based and bugger-based algorithms will become more seamless.
Neutral networks also fine-tunes itself over time thanks to a system created based on rewards and penalties. This will then allow streaming services to customize content for more buffering or resolution issues. Companies will be able to better predict if their viewer is about to hit with a rash of bad connectivity, and find the right streaming solution.