Exemplar-Based Sparse Representations for Noise Robust Automatic Speech Recognition

By Jort Florent Gemmeke, Tuomas Virtanen, Antti Hurmalainen. IEEE Transactions on Audio, Speech, and Language Processing, 2011

380 citations recorded by the source corpus.

Subfields: Mechanics of Materials, Signal Processing

View Paper: https://www.semanticscholar.org/paper/e68001a21a210bdd403a00b55d4d3633c4b1ea58

How researchers see this paper

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

In Mechanics of Materials

1. Li Deng
2. Naif Alajlan

In Signal Processing

1. Jinyu Li
2. Dong Yu
3. Yifan Gong
4. Li Deng
5. Xiao Li
6. Zixing Zhang
7. Zhiheng Huang
8. Björn Schuller

Citations in the corpus

• Land-Use Classification With Compressive Sensing Multifeature Fusion (2015, background): “, the compressive sensing (CS) theory, which has emerged as a mainstream in information theory in recent years [7], [8], and ever since, it has been adopted in various applications mainly related to computer vision [9], remote sensing [10], speech recognition [11], source separation [12], and assist”
• An Overview of Noise-Robust Automatic Speech Recognition (2014, methodology): “This is a source separation approach, and non-negative matrix factorization (NMF) [166] has been shown to be a very successful method [167], [168], and can directly bene fi t noise-robust ASR [163]–[165], [169].”
• Learning Noise-Invariant Representations for Robust Speech Recognition (2018, background): “[31] demonstrates that modeling speech as a linear combination of exemplars results in noise-robust ASR models.”
• Multi-view Recurrent Neural Acoustic Word Embeddings (2016, neutral): “Such embeddings could be useful for tasks like spoken term detection (Fiscus et al., 2007), spoken query-by-example search (Anguera et al., 2014), or even speech recognition using a whole-word approach (Gemmeke et al., 2011; Bengio & Heigold, 2014).”
• Machine Learning Paradigms for Speech Recognition: An Overview (2013, background): “In the studies of [265], [266], various uncertainty measures are developed to characterize the expected accuracy of a sparse imputation, an exemplar-based reconstruction method based on representing segments of the noisy speech signal as linear combinations of as few clean speech example segments as”
• Discriminative piecewise linear transformation based on deep learning for noise robust automatic speech recognition (2013, methodology): “Feature enhancement approaches, such as SPLICE [2], Denoising AutoEncoder (DAE) [3], and so on [4–6] are techniques that can be performed on the front-end for this aim.”
• Deep Learning for Environmentally Robust Speech Recognition (2017, positive): “ionaries of speech and noise is by training them separately on the clean speech and noise sounds [64], [65]. More advanced variations include exemplar-based approaches to obtaining large dictionaries [66], and semi-supervised approaches to releasing the requirement of both speech and noise for dicti”
• Speech Dereverberation Using Non-Negative Convolutive Transfer Function and Spectro-Temporal Modeling (2016, background): “Hence, since individual frequency bins are processed independently, these methods completely ignore the spectral structure of the speech signal.”