Enhancement and bandwidth compression of noisy speech

By Jae S. Lim, A. Oppenheim. 1979

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

1. Jinyu Li
2. Li Deng
3. Dong Yu
4. Deliang Wang
5. Yossi Adi
6. Alexandre D'efossez
7. Gabriel Synnaeve
8. Yoshua Bengio

Citations in the corpus

• Real Time Speech Enhancement in the Waveform Domain (2020, background): “Decades of work in speech enhancement showed feasible solutions which estimated the noise model and used it to recover noise-deducted speech [4, 5].”
• Speech Enhancement in Multiple-Noise Conditions Using Deep Neural Networks (2016, methodology): “The wiener filtering [4] based approach was proposed in the same year.”
• Deep Residual-Dense Lattice Network for Speech Enhancement (2020, methodology): “Here, RDL-Nets were used to estimate the a priori SNR for the SRWF approach and the MMSE-LSA estimator.”
• On the optimality of ideal binary time-frequency masks (2008, background): “It is worth pointing out that the SNR metric does not correlate directly with speech intelligibility or quality when the signal is speech (Lim and Oppenheim, 1979).”
• An Overview of Noise-Robust Automatic Speech Recognition (2014, background): “Another example is the famous Wiener fi lter, proposed as early as 1979 [106] to improve the performance on noisy speech.”
• A Regression Approach to Speech Enhancement Based on Deep Neural Networks (2015, methodology): “, [3], [4]) is that the resulting enhanced speech often suffers from an annoying artifact called “musical noise” [5].”
• A Fully Convolutional Neural Network for Speech Enhancement (2016, background): “work 1. INTRODUCTION Denoising speech signals has been a long standing problem. Decades of works showed feasible solutions which estimated the noise model and used it to recover noise-deducted speech [1, 2, 3, 4, 5]. Nonetheless, estimating the model for a babble noise, which is encountered when a c”
• A joint resonance frequency estimation and in-band noise reduction method for enhancing the detectability of bearing fault signals (2008, methodology): “Furthermore, a good estimate of jN̂ðoÞj(2) can be found by analyzing the intervals where speech is not present [15].”
• The importance of phase in speech enhancement (2011, methodology): “Traditional AMS-based speech enhancement methods (e.g., Lim and Oppenheim, 1979; Berouti et al., 1979; Ephraim and Malah, 1984; Ephraim et al., 1985; Martin, 1994; Sim et al., 1998; Virag, 1999), modify only the noisy magnitude spectrum, while keeping the noisy phase spectrum unchanged.”
• Speech enhancement with weighted denoising auto-encoder (2013, methodology): “In the same year, speech enhancement method with frequency domain Wiener filter [2] was proposed by J.”