Geodesic
Overview of current open solutions in the field of speech recognition
News of the Kabardin-Balkar scientific center of RAS, no. 6 (2022), pp. 127-133.

Voir la notice de l'article provenant de la source Math-Net.Ru

The purpose of this work is to review the most successful open solutions in the field of speech recognition and also considers the processes of speech recognition and the possibilities of their practical use. The paper presents classical solutions based on recurrent neural networks, as well as more modern ones, which use convolutional neural networks as a basis to remove noise and reduce dimensionality, and transformers that allow to memorize the context and work with the semantic meaning of sequences, regardless of time.
Keywords: artificial intelligence, speech recognition, neural networks, natural language processing, convolutional neural networks, recurrent neural networks, transformers. 
@article{IZKAB_2022_6_a11,
     author = {K. V. Nalchadzhi},
     title = {Overview of current open solutions in the field of speech recognition},
     journal = {News of the Kabardin-Balkar scientific center of RAS},
     pages = {127--133},
     publisher = {mathdoc},
     number = {6},
     year = {2022},
     language = {ru},
     url = {http://geodesic.mathdoc.fr/item/IZKAB_2022_6_a11/}
}
TY  - JOUR
AU  - K. V. Nalchadzhi
TI  - Overview of current open solutions in the field of speech recognition
JO  - News of the Kabardin-Balkar scientific center of RAS
PY  - 2022
SP  - 127
EP  - 133
IS  - 6
PB  - mathdoc
UR  - http://geodesic.mathdoc.fr/item/IZKAB_2022_6_a11/
LA  - ru
ID  - IZKAB_2022_6_a11
ER  - 
%0 Journal Article
%A K. V. Nalchadzhi
%T Overview of current open solutions in the field of speech recognition
%J News of the Kabardin-Balkar scientific center of RAS
%D 2022
%P 127-133
%N 6
%I mathdoc
%U http://geodesic.mathdoc.fr/item/IZKAB_2022_6_a11/
%G ru
%F IZKAB_2022_6_a11
K. V. Nalchadzhi. Overview of current open solutions in the field of speech recognition. News of the Kabardin-Balkar scientific center of RAS, no. 6 (2022), pp. 127-133. http://geodesic.mathdoc.fr/item/IZKAB_2022_6_a11/

[1] Hemant Yadav, Sunayana Sitaram [et al], “A Survey of Multilingual Models For Automatic Speech Recognition”, 2022, arXiv: abs/2202.12576

[2] Awni Hannun, Carl Case, Jared Casper [et al], Deep Deep Speech: Scaling up end-to-end speech recognition, 2007, arXiv: abs/1412.5567v2 | MR

[3] Roger Grosse, Helen Kwong, Andrew Y. Ng [et al], Shift-Invariance Sparse Coding for Audio Classification, 2012, arXiv: abs/1206.5241

[4] Awni Y. Hannun, Daniel Jurafsky, Andrew Y. Ng [et al], First-Pass Large Vocabulary Continuous Speech Recognition using Bi-Directional Recurrent DNNs, 2014, arXiv: abs/1408.2873

[5] Anmol Gulati, James Qin, Chung-Cheng Chiu [et al], Conformer: Convolution-augmented Transformer for Speech Recognition, 2020, arXiv: abs/2005.08100

[6] Andrew L. Maas, Peng Qi, Ziang Xie [et al], Building DNN Acoustic Models for Large Vocabulary Speech Recognition, 2014, arXiv: abs/1406.7806

[7] Kaitao Song, Xu Tan, Di He, Jianfeng Lu [et al], “Double Path Networks for Sequence to Sequence Learning”, In Proceedings of the 27th International Conference on Computational Linguistics, 2018, 3064–3074, arXiv: abs/1806.04856

[8] Dario Amodei, Rishita Anubhai, Eric Battenberg [et al], Deep Speech 2: End-to-End Speech Recognition in English and Mandarin, 2015, arXiv: abs/1512.02595

[9] Tianxiao Shen, Myle Ott, Michael Auli [et al], Mixture Models for Diverse Machine Translation: Tricks of the Trade, 2019, arXiv: abs/1902.07816

[10] Neil Zeghidour, Qiantong Xu, Vitaliy Liptchinsky [et al], Wav2Letter++: Fully Convolutional Speech Recognition, 2018, arXiv: abs/1812.06864

[11] Noam Shazeer, Azalia Mirhoseini, Krzysztof Maziarz [et al], Outrageously Large Neural Networks: The Sparsely-Gated Mixture-of-Experts Layer, 2017, arXiv: abs/1701.06538

[12] Shashi Narayan, Shay B. Cohen, Mirella Lapata [et al], Don't Give Me the Details, Just the Summary! Topic-Aware Convolutional Neural Networks for Extreme Summarization, 2018, arXiv: abs/1808.08745

[13] Myle Ott, Sergey Edunov, Alexei Baevski [et al], Fairseq: A Fast, Extensible Toolkit for Sequence Modeling, 2019, arXiv: abs/1904.01038

[14] An Analysis of Neural Language Modeling at Multiple Scales, Stephen Merity, Nitish Shirish Keskar, Richard Socher [et al], 2018, arXiv: abs/1803.08240

[15] Sebastian Gehrmann, Yuntian Deng, Alexander M. Rush [et al], Bottom Up Abstractive Summarization, 2018, arXiv: abs/1808.10792

[16] Shamil Chollampatt, Ng. Hwee Tou, A Multilayer Convolutional Encoder Decoder Neural Network for Grammatical Error Correction, 2018, arXiv: abs/1801.08831

[17] Steffen Schneider, Alexei Baevski, Ronan Collobert [et al], wav2vec: Unsupervised Pretraining for Speech Recognition, 2019, arXiv: abs/1904.05862

[18] Gabriel Synnaeve, Qiantong Xu, Jacob Kahn [et al], End-to-end ASR: from Supervised to Semi-Supervised Learning with Modern Architectures, 2019, arXiv: abs/1911.08460

[19] Jayadev Billa, Improving low-resource ASR performance with untranscribed out-of-domain data, 2021, arXiv: abs/2106.01227

[20] Wei-Ning Hsu, Benjamin Bolte, Yao-Hung Hubert Tsai [et al], Self-Supervised Speech Representation Learning by Masked Prediction of Hidden Units, HuBERT, 2021, arXiv: abs/2106.07447

[21] Tahir Javed, Sumanth Doddapaneni, Abhigyan Raman [et al], Towards Building ASR Systems for the Next Billion Users, 2021, arXiv: abs/2111.03945