🤖 AI Summary
This study addresses the challenge of contamination in stereo audio recordings caused by real-world train noise and environmental acoustic interference. To this end, it proposes a multi-reference recursive least squares (RLS) adaptive noise cancellation method that operates without requiring a clean reference signal. The approach leverages stereo reference signals originating from the same noise source, employing per-channel 30th-order adaptive filters, a 15th-order anti-causal structure, a forgetting factor of 0.999, and an FIR low-pass post-filter to effectively model and suppress interference components under complex propagation conditions. Experimental results demonstrate that the method reduces the correlation between residual output and reference signals to 0.011–0.016, achieving a correlation ratio attenuation of 30.6–34.1 dB and an RMS reduction of 1.8–4.8 dB. This work significantly enhances audio quality in the absence of ground-truth clean data and represents the first successful application of efficient multichannel noise cancellation to real train noise.
📝 Abstract
Reference-based adaptive interference cancellation is evaluated for stereo audio recordings corrupted by real train noise and environmental background. The observed signal is modeled as a clean stereo program contaminated by an additive disturbance generated by an external acoustic source through unknown propagation paths. A second stereo recording, representing another filtered observation of the same physical noise source, is used as the reference input of a multi-reference recursive least-squares (RLS) estimator. The estimated train-interference component is subtracted from the noisy audio and followed by a finite-impulse-response low-pass postfilter. Three 74.01 s real audio sequences sampled at 11.025 kHz are processed under identical algorithmic parameters. Since clean ground truth is not available, performance is assessed with no-reference indicators: waveform behavior, Welch spectral estimates, RMS change, and residual normalized correlation with the reference. With 30 taps per reference channel, 15 anti-causal taps, and forgetting factor 0.999, the maximum reference correlation is reduced from 0.386--0.832 before processing to 0.011--0.016 after processing. The corresponding correlation-ratio reduction is approximately 30.6--34.1 dB, while the output RMS decreases by 1.8--4.8 dB depending on section and stereo channel. The results demonstrate that real train interference, including environmental acoustic effects, can be substantially attenuated when a correlated reference recording is available.