Noise-canceling headphones have become a popular choice for many individuals seeking to block out unwanted background noise. However, the ability to choose which sounds to filter through the headphones has been limited. Recognizing the need for more control over sound selection, a team of researchers from the University of Washington has developed deep-learning algorithms that allow users to pick and prioritize the sounds they want to hear in real time.
The system, dubbed “semantic hearing,” involves streaming audio captured by the headphones to a connected smartphone. The smartphone then cancels out all environmental sounds, creating a clean audio experience for the user. Through either voice commands or a dedicated smartphone app, wearers can select from a range of 20 sound classes, such as sirens, baby cries, speech, vacuum cleaners, and bird chirps. Only the chosen sounds will be played through the headphones, allowing for a personalized audio experience.
The team presented their research at the UIST ’23 conference in San Francisco on November 1. They aim to release a commercial version of the system in the future.
Senior author Shyam Gollakota, a professor in the Paul G. Allen School of Computer Science & Engineering at the University of Washington, highlighted the challenge of extracting specific sounds from a complex environment in real time. Previous noise-canceling headphones have not achieved this level of intelligent filtering. The semantic hearing system has to process sounds in less than a hundredth of a second to ensure synchronization with the wearer’s visual senses.
To meet this demanding requirement, the system processes the sounds on a connected smartphone rather than relying on cloud servers, allowing for real-time processing. Additionally, the system preserves the delays and spatial cues that are important for perceiving sound in its original environment. This ensures that wearers can still effectively navigate and understand their surroundings while receiving a customized audio experience.
The system underwent testing in various environments, including offices, streets, and parks. It successfully extracted target sounds such as sirens, bird chirps, and alarms while removing other background noise. Participants who tested the system reported improved audio quality compared to the original recordings.
However, the system encountered some difficulties when distinguishing between sounds that share similar properties, such as vocal music and human speech. The researchers believe that training the models on more real-world data could enhance the system’s ability to accurately differentiate between these sounds.
The development of AI-powered noise-canceling headphones with customizable sound selection opens up new possibilities for individuals seeking a more tailored audio experience. Whether it’s blocking out unwanted car horns while working indoors or selectively hearing bird chirps during a walk along busy streets, wearers can now have greater control over the sounds they choose to hear.
As the researchers continue to refine and commercialize the system, users can look forward to a new era of personalized audio technology that seamlessly integrates with their daily lives. Whether for work or leisure, these intelligent headphones have the potential to enhance the sound experience for users in a wide range of environments.
*Note:
- Source: Coherent Market Insights, Public sources, Desk research
- We have leveraged AI tools to mine information and compile it
Ravina Pandya, Content Writer, has a strong foothold in the market research industry. She specializes in writing well-researched articles from different industries, including food and beverages, information and technology, healthcare, chemical and materials, etc. With an MBA in E-commerce, she has an expertise in SEO-optimized content that resonates with industry professionals.