New framework makes noise-cancelling earbuds robust to user fit variations
A smarter algorithm keeps noise cancellation consistent even when earbuds shift or users change.
Spatially selective active noise control (SSANC) hearables aim to cancel noise from specific directions while preserving desired speech from others. A key challenge is that the secondary path — the acoustic path from the earbud's loudspeaker to its inner error microphone — varies dramatically across users and even between different insertions. Existing systems that assume a fixed, accurate secondary path estimate often suffer degraded performance or instability in real-world use.
Tong Xiao and colleagues from Oldenburg University address this with a robust soft-constrained optimization framework. Instead of relying on a single path estimate, they compute a control filter that minimizes average cost over a set of secondary path measurements derived from actual human data. Their simulations and real-time experiments demonstrate that while the mean noise reduction is slightly lower than in the perfectly matched case, the variability of performance across different paths is greatly reduced. This trade-off makes the approach practical for mass-produced hearables where individualized calibration is infeasible. The work was submitted to IWAENC 2026.
- Proposed robust soft-constrained optimization for SSANC hearables averages cost over multiple secondary path measurements.
- Real-time experiments show mean noise reduction slightly lower but performance spread significantly narrowed under path mismatch.
- Addresses practical issue of secondary path variation across users and earbud fits without requiring individual calibration.
Why It Matters
Makes active noise cancellation in earbuds consistently effective across different users and wearing conditions.