New AI Cleans Up Background Noise in Calls Without Lag
This could make your next phone call sound like a studio recording.
GTCRN-SS is a two-stage time-frequency separator that adapts the compact GTCRN enhancement backbone to fixed-output speech separation. It combines internal residual allocation, reconstruction-conditioned refinement, and frame-aligned recurrent-state transfer; its complex-mask estimator assigns non-target content to an unreturned residual branch, while the complementary speech estimate conditions a second stage whose final masks remain anchored to the original mixture.
A lightweight low-rank adapter transfers causal recurrent evidence between recursive executions without adding another recurrent memory network. The neural graph excludes future feature frames and incurs only the fixed analysis look-ahead of the centered short-time Fourier transform.
On noisy Libri2Mix, GTCRN-SS achieves 10.1 dB SDRi and 9.5 dB SI-SNRi with fewer than one million parameters and a low computational footprint. Independently trained ablations confirm the benefits of recurrent-state transfer and reconstruction conditioning, and corpus-level diagnostics show that the internal branch primarily captures residual rather than target-speech content. These results establish GTCRN-SS as an
- GTCRN-SS is a small AI that removes background noise from speech in real time, without delays.
- It uses less than a million parameters, so it can run on phones and earbuds without heavy computing.
- In tests, it improved speech clarity by about 10 decibels, making voices easier to understand.
Why It Matters
Clearer calls in noisy places, better online meetings, and smarter hearing aids without lag.