摘要 |
A noisy audio signal, with user input device noise, is received. Particular frames in the audio signal that are corrupted by user input device noise are identified and removed. The removed audio data is then reconstructed to obtain a clean audio signal.
|