Joint noise adaptive training for robust automatic speech recognition

By Arun Narayanan, Deliang Wang. 2014

110 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

Exploring Speech Enhancement with Generative Adversarial Networks for Robust Speech Recognition (2017, neutral): “Furthermore, FSEGAN degraded performance for ASR-MTR, consistent with observations in [33].”
Batch-normalized joint training for DNN-based distant speech recognition (2016, background): “Although joint training for speech recognition is still an under-explored research direction, such a paradigm is progressively gaining more attention and some interesting works in the ﬁeld have been recently published [29, 30, 31, 32, 33, 34, 35].”
Deep Learning for Environmentally Robust Speech Recognition (2017, methodology): “Furthermore, the work done in Narayanan and Wang (2014) even left out the pre-training process and directly concatenated a DNN-based speech separation front-end and a DNN-based AM back-end to build a large neural network and jointly adjusted all of the weights.”
The Pytorch-kaldi Speech Recognition Toolkit (2018, unknown): “It is also possible to employ models based on an ecosystem of neural networks performing different tasks, as done in the context of joint training between speech enhancement and speech recognition [28, 29] or in the context of the recently-proposed cooperative networks of deep neural networks [30].”
A deep neural network for time-domain signal reconstruction (2015, background): “(4) By expressing the inverse DFT operation in matrix form and plugging in Eq.”
Improving Robustness of Deep Neural Network Acoustic Models via Speech Separation and Joint Adaptive Training (2015, methodology): “It was shown in [30] that instead of using such an estimate, speech separation can be used to obtain a more accurate estimate of noise.”
Text-informed speech enhancement with deep neural networks (2015, unknown): “…the text features in a speech enhancement framework previously, DNN may al-Figure As shown in many literatures including TTS [12, 13], automatic speech recognition (ASR) [20–22] and enhancement research [23], the nature of DNN allows us to utilize disparate features jointly in an inference stage.”
Deep neural network based spectral feature mapping for robust speech recognition (2015, methodology): “It is interesting to train the two DNNs jointly, similar to the mask estimation in [11, 24].”
Spatial diffuseness features for DNN-based speech recognition in noisy and reverberant environments (2014, methodology): “This has been evaluated for stationary noise estimates [3] and noise-estimates derived from time-frequency masking [9], but may in principle also be used for noise estimates obtained from spatial processing.”
An experimental study on joint modeling of mixed-bandwidth data via deep neural networks for robust speech recognition (2016, methodology): “Recently, DNN has been used as a regression model to enhance noisy speech [16] followed by robust ASR [17], [18], [19].”
Learning Spectral Mapping for Speech Dereverberation and Denoising (2015, background): “We mention that these ASR experiments aim to show the promise of DNN based dereverberation and denoising rather than reach the state-of-the-art results (see [25]).”
A network of deep neural networks for Distant Speech Recognition (2017, methodology): “Within the DNN framework, the adoption of a joint training approach between speech enhancement and speech recognition DNNs has recently been proposed to mitigate the lack-of-matching issue [17–24].”