| Summary: | 碩士 === 國立清華大學 === 動力機械工程學系 === 103 === Acoustic echo is one of the key issues that must be addressed in telecommunication. In this work, an evolutionary exposition is given in regard to the enhancing strategies for acoustic echo cancellers (AEC). A fixed beamformer (FBF) is utilized to focus on the near-end speaker while suppressing the echo from the far end. In reality, the array steering vector could differ from the ideal freefield plane wave model. Therefore, an experimental procedure is developed to interpolate a practical array model from the measured frequency responses. Subband (SB) filtering with is exploited to accelerate the cancellation process. Generalized Sidelobe Canceller (GSC) comprised of a FBF and an adaptive blocking module is combined with AEC to maximize cancellation performance. Another major enhancement is an internal iteration (IIT) procedure that enables coefficients update in the adaptive SB filters within a sample time. Convergence is significantly improved in terms of speed and residual error by using the iterative scheme. Objective tests in terms of Echo Return Loss Enhancement (ERLE), Perceptual Evaluation of Speech Quality (PESQ) and recognition rate in Automatic Speech Recognition (ASR) and subjective listening tests are conducted to validate the proposed enhanced AEC approaches. The results show that the GSC-SB-AEC-IIT approach has attained the highest ERLE with the best speech quality, even in double-talk scenarios.
|
|---|
