Hybrid acoustic models for distant and multichannel large vocabulary speech recognition

By Pawel Swietojanski, Arnab Ghoshal, S. Renals. 2013

112 citations recorded by the source corpus.

Subfields: Artificial Intelligence, Model Architectures, Speech & Audio, 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 Model Architectures

In Speech & Audio

In Signal Processing

Citations in the corpus

Highway long short-term memory RNNS for distant speech recognition (2015, background): “In addition, we used dropout as the way to control the highway connections which turns out to be critical for DSR. 1The tools and scripts used to produce the results reported in this paper are publicly available as part of the CNTK toolkit[15], and anyone with access to the data should be able to re”
The DIRHA-ENGLISH corpus and related tasks for distant-speech recognition in domestic environments (2015, background): “…at multi-microphone front-end processing level in order to feed ASR with an enhanced speech input [4, 5, 6, 7, 8], the performance loss observed from close-talking to distant-speech remains quite critical, even when the most advanced DNN-based backend frameworks are adopted [9, 10, 11, 12, 13, 14].”
Using neural network front-ends on far field multiple microphones based speech recognition (2014, background): “Similar work but with a different focus and system setup is referred in [17].”
Realistic Multi-Microphone Data Simulation for Distant Speech Recognition (2016, background): “…[3], source separation [4], speech dereverberation [5], speaker localization [6], acoustic event detection [7] as well as acoustic modeling [8, 9, 10, 11, 12], DSR still exhibits a lack of robustness, especially when adverse acoustic conditions originated by non-stationary noises and…”
A network of deep neural networks for Distant Speech Recognition (2017, background): “…the recent success of deep learning has not only largely contributed to the substantial improvement of the speech recognition part of a DSR system [12–14], but has also enabled the development of competitive DNN-based speech enhancement solutions [15,16], making an effective integration be-tween…”
Far-Field ASR Without Parallel Data (2016, methodology): “However, unlike in [8], we perform speaker-adaptive training of the HMM-GMM systems for all the three tasks, as we found the alignments from SAT HMMGMM systems to be beneficial for neural network training on all three tasks.”
Deep Learning for Environmentally Robust Speech Recognition (2017, methodology): “In Swietojanski et al. (2013), the individual features extracted from each microphone channel are concatenated as a long single feature vector and fed into a DNN for AM.”
A Study of Enhancement, Augmentation, and Autoencoder Methods for Domain Adaptation in Distant Speech Recognition (2018, background): “Deep neural networks work particularly well with multi-condition training due to the large model capacity [7, 10].”
Convolutional Neural Networks for Distant Speech Recognition (2014, background): “Similarly, it is possible for each convolutional band to have a separate learnable bias parameter instead of the biases only being shared across bands [25], [26].”
Batch-normalized joint training for DNN-based distant speech recognition (2016, background): “Nevertheless, the recent success of deep learning has not only largely contributed to the substantial improvement of the speech recognition part of a DSR system [19, 20, 21, 22, 23, 24, 25], but has also enabled the de-a r X i v : 1703 .”
Speech acoustic modeling from raw multichannel waveforms (2015, methodology): “In the multichannel case (Table 2) we compare the waveform CNN against stacked mel-tb features similar to [13, 14] (but without pooling across channels) and against mel-tb features computed on the output of a delay-and-sum beamformer.”
Contaminated speech training methods for robust DNN-HMM distant speech recognition (2017, background): “Concerning the latter field, the research community is currently experiencing a revolution due to the introduction of Deep Neural Networks (DNNs) [13], which have consistently outperformed previous Gaussian Mixture Models (GMMs) for both close [14] and distant-talking scenarios [15, 16, 17, 18, 19].”
The NTT CHiME-3 system: Advances in speech enhancement and recognition for mobile multi-microphone devices (2015, background): “Note that the effect of multi-microphone training was examined previously in [16] for meeting transcription.”
Spatial diffuseness features for DNN-based speech recognition in noisy and reverberant environments (2014, background): “In [10] and [11], feature vectors from multiple microphones are concatenated to form the input of a DNN-based acoustic model, however, no spatial phase information is exploited.”
Factorized Hidden Layer Adaptation for Deep Neural Network Based Acoustic Modeling (2016, unknown): “These matrices and vectors are learned adaptively with the speaker representations, which are initialized using the speaker i-vectors and later updated using unsupervised adaptation.”
Single-Channel Multi-talker Speech Recognition with Permutation Invariant Training (2017, methodology): “These results are comparable with that reported by others [41] even though we did not use adapted fMLLR feature.”