Noise canceling headphones have become very good at creating a barrier that filters out outside sounds. This is possible, Basically, thanks to microphones that pick up and cancel out the environment With other frequencies. But this cancels out all the voices, and does not allow us to distinguish between those that are important to us. At least that’s how it has been so far.
A team from the University of Washington has developed an artificial intelligence system that can It allows the user wearing headphones to look at the person speaking to “record” them.. The system, called Target Speech Hearing, cancels out all other sounds in the environment and reproduces only the recorded speaker’s voice in real time, even when the listener moves through noisy spaces and is no longer looking at the sound emitter. The team is led by Shyam Gollakota. Present your findings In ACM CHI Conference on Human Factors in Computing Systems. The code for the proof-of-concept device is available for others to develop, but it is not commercially available at this time.
“When thinking about AI as being based on chatbots The web that answers questions – Golakota explains at a permit-. But in this project We are developing artificial intelligence to modify perceptionalsoHear someone To use headphones, according to your preference. With our devices, you can now hear one speaker clearly, even if it istoIn a noisy environment with many other people talking.
To use the system, a person wearing standard headphones with microphones presses a button while doing so Point your head toward someone speaking for 3 to 5 seconds. The sound waves from the speaker’s voice must reach the microphones on both sides of the headphones at the same time; There is a margin of error of 16 degrees. The headphones send that signal to an onboard computer, where the device’s machine learning software learns the vocal patterns of the intended speaker.
The system picks up that person’s voice and captures it It continues to play for the listener, even if they both move. The system’s ability to focus on recorded audio improves as the speaker continues to speak, providing the system with more training data. Taking into account that our voice can be used as a fingerprint, what the system basically does is associate the sound with a specific pattern of sound waves and search for it permanently.
The team tested their system on 21 people and evaluated… The clarity of recorded speaker sound is almost twice that of unfiltered sound on average. Currently, the TSH system can only record one speaker at a time, and can only record a speaker when there is no other strong sound coming from the same direction as the target speaker’s voice. If the user is not satisfied with the sound quality, they can make another recording on the speaker to improve clarity. The team is working on expanding the system to include headphones and earphones in the future.
“Beer enthusiast. Subtly charming alcohol junkie. Wannabe internet buff. Typical pop culture lover.”