Exploring The Robustness Of Features And Enhancement On Speech Recognition Systems In Highly-reverberant Real Environments

Abstract

This paper evaluates the robustness of a DNN-HMM-based speech recognition system in highly-reverberant real environments using the HRRE database. The performance of locally-normalized filter bank (LNFB) and Mel filter bank (MelFB) features in combination with Non-negative Matrix Factorization (NMF), Suppression of Slowly-varying components and the Falling edge (SSF) and Weighted Prediction Error (WPE) enhancement methods are discussed and evaluated. Two training conditions were considered: clean and reverberated (Reverb). With Reverb training the use of WPE and LNFB provides WERs that are 3% and 20% lower in average than SSF and NMF, respectively. WPE and MelFB provides WERs that are 11% and 24% lower in average than SSF and NMF, respectively. With clean training, which represents a significant mismatch between testing and training conditions, LNFB features clearly outperform MelFB features. The results show that different types of training, parametrization, and enhancement techniques

Stats

• citations0
• S2 citations—
• github stars0
• HF likes0
• heat score0.00
• arxiv keynovoa2018exploring

Related papers

• On Combining Features For Single-channel Robust Speech Recognition In Reverberant Environments (2019)0.00
• An Improved Dnn-based Spectral Feature Mapping That Removes Noise And Reverberation For Robust Automatic Speech Recognition (2018)0.00
• Neural Network-augmented Kalman Filtering For Robust Online Speech Dereverberation In Noisy Reverberant Environments (2022)0.00
• On The Use Of DNN Autoencoder For Robust Speaker Recognition (2018)0.00
• Run-time Adaptation Of Neural Beamforming For Robust Speech Dereverberation And Denoising (2024)0.00
• Deep Learning Based Dereverberation Of Temporal Envelopesfor Robust Speech Recognition (2020)5.84
• Ensemble Of Jointly Trained Deep Neural Network-based Acoustic Models For Reverberant Speech Recognition (2016)0.00
• Speech Enhancement With Wide Residual Networks In Reverberant Environments (2019)0.00