Long Short-Term Memory for Speaker Generalization in Supervised Speech Separation

By Jitong Chen, Deliang Wang. 2016

251 citations recorded by the source corpus.

Subfields: Artificial Intelligence, Signal Processing

How researchers see this paper

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

In Artificial Intelligence

In Signal Processing

Citations in the corpus

Speech Denoising with Deep Feature Losses (2018, background): “However, most pipelines still operate in the spectrogram domain [4, 5, 6, 7, 8, 9, 10, 11].”
Fullsubnet: A Full-Band and Sub-Band Fusion Model for Real-Time Single-Channel Speech Enhancement (2020, unknown): “The full-band model’s architecture, i.e., a stack of LSTM layers, is actually widely used for speech enhancement, such as in [6,26].”
DeepMMSE: A Deep Learning Approach to MMSE-Based Noise Power Spectral Density Estimation (2020, background): “LSTM-IRM is an LSTM network that estimates the ideal ratio mask (IRM) [57], and Xu2017 is a multi-layer perceptron (MLP) clean speech spectrum estimator that incorporates multi-objective learning and ideal binary mask (IBM) post-processing (available from https://github.com/…”
Gated Residual Networks with Dilated Convolutions for Supervised Speech Separation (2018, unknown): “We find that the proposed GRN leads to consistently better performance and higher computational efficiency than the LSTM model in [6].”
A New Framework for CNN-Based Speech Enhancement in the Time Domain (2019, methodology): “For the enhancement task on this dataset, the importance of future and past context was first established in [28].”
Supervised Speech Separation Based on Deep Learning: An Overview (2017, unknown): “STOI improvements of a feedforward DNN and a RNN with LSTM (from [20]).”
Deep Residual-Dense Lattice Network for Speech Enhancement (2020, methodology): “Recurrent neural networks (RNNs) employing long short-term memory (LSTM) cells provided a higher performance at the cost of parameter inef-ﬁciency and extensive training times (Chen and Wang 2017).”
A Convolutional Recurrent Neural Network for Real-Time Speech Enhancement (2018, methodology): “[10] proposed an RNN with four hidden LSTM layers to deal with speaker generalization of noise-independent models.”
Speech Enhancement Using Bayesian Wavenet (2017, background): “A series of works [5,6], applied different deep learning architectures to predict ideal ratio masks.”
Joint Phoneme Alignment and Text-Informed Speech Separation on Highly Corrupted Speech (2020, methodology): “It resembles the speech separation model in [4] which is a four-layer LSTM-RNN with a linear output layer.”
Learning Complex Spectral Mapping with Gated Convolutional Recurrent Networks for Monaural Speech Enhancement (2019, background): “[1] proposed an RNN with four hidden LSTM layers to address speaker generalization of noise-independent models.”