ClearBuds
Wireless Binaural Earbuds for Learning-Based Speech Enhancement
ClearBuds are the first wireless earbuds that clear up calls using deep learning. With the compact form factor truly wireless stereo earbuds provide, more and more people are taking calls on-the-go. ClearBuds can help to provide people with the ability to take calls in noisy environments. Our neural network runs completely on an iPhone, allowing you to suppress unwanted noises while taking phone calls on the go. ClearBuds bridges state-of-the-art deep learning for blind audio source separation and in-ear mobile systems by making two key technical contributions: 1) a new wireless earbud design capable of operating as a synchronized, binaural microphone array, and 2) a lightweight dual-channel speech enhancement neural network that runs on a mobile device. Results show that our wireless earbuds achieve a synchronization error less than 64 microseconds and our network has a runtime of 21.4 ms on an accompanying mobile phone.
ClearBuds were born out of necessity. In addition to being lead co-authors on this paper, Maruchi, Vivek and I were roommates when the pandemic lockdown started. Like many others, we found ourselves taking many calls in close quarters. Between kitchen, construction, and conversations, our house was a noisy environment for taking these calls. With collective experience across hardware, networking, and machine learning, we thought we could do something about it, and arrived on the idea for ClearBuds pretty quickly. We ultimately presented our paper and ran a real-time demo at ACM Mobisys 2022. The work earned runner-up for best demo at the conference.
You may find more information about ClearBuds here, including links to the paper here: clearbuds.cs.washington.edu.
ClearBuds were born out of necessity. In addition to being lead co-authors on this paper, Maruchi, Vivek and I were roommates when the pandemic lockdown started. Like many others, we found ourselves taking many calls in close quarters. Between kitchen, construction, and conversations, our house was a noisy environment for taking these calls. With collective experience across hardware, networking, and machine learning, we thought we could do something about it, and arrived on the idea for ClearBuds pretty quickly. We ultimately presented our paper and ran a real-time demo at ACM Mobisys 2022. The work earned runner-up for best demo at the conference.
You may find more information about ClearBuds here, including links to the paper here: clearbuds.cs.washington.edu.
Citation
Ishan Chatterjee, Maruchi Kim, Vivek Jayaram, Shyamnath Gollakota, Ira Kemelmacher, Shwetak Patel, and Steven M. Seitz. 2022. ClearBuds: wireless binaural earbuds for learning-based speech enhancement. In Proceedings of the 20th Annual International Conference on Mobile Systems, Applications and Services (MobiSys '22). Association for Computing Machinery, New York, NY, USA, 384–396. https://doi.org/10.1145/3498361.3538933
Ishan Chatterjee, Maruchi Kim, Vivek Jayaram, Shyamnath Gollakota, Ira Kemelmacher, Shwetak Patel, and Steven M. Seitz. 2022. ClearBuds: wireless binaural earbuds for learning-based speech enhancement. In Proceedings of the 20th Annual International Conference on Mobile Systems, Applications and Services (MobiSys '22). Association for Computing Machinery, New York, NY, USA, 384–396. https://doi.org/10.1145/3498361.3538933
© Ishan Chatterjee 2023