To develop an AI/ML-enabled adaptive noise cancellation (ANC) system that effectively suppresses stationary, non-stationary, and impulsive defence noises while maintaining high speech intelligibility and real-time performance on embedded hardware.

Source summary

The source proposes AI/ML-assisted noise suppression for intelligible real-time communication in impulsive and changing high-noise environments. It asks for noisy/clean speech datasets, model training, an edge-deployable inference engine and microphone/headset demonstration; an additional adaptive filter is optional. Requested evaluation targets are SNR above 15 dB, STOI above 0.85 and PESQ above 2.5 at low latency. This is a high-level communications description, not evidence those targets have been achieved.