Speech Recognition with No Speech or with Noisy Speech

Krishna, Gautam; Tran, Co; Yu, Jianguo; Tewfik, A.H.

doi:10.1109/icassp.2019.8683453

Cited by 56 publications

(61 citation statements)

References 30 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In [1] we have demonstrated that EEG sensors T7 and T8 features contributed most towards ASR performance. Table VI shows the CTC model test time results when we trained the model using EEG features from only T7 and T8 sensors on the most noisy data set B.…”

Section: Resultsmentioning

confidence: 86%

“…EEGlab's [18] Independent component analysis (ICA) toolbox was used to remove other biological signal artifacts like electrocardiography (ECG), electromyography (EMG), electrooculography (EOG) etc from the EEG signals. We extracted five statistical features for EEG, namely root mean square, zero crossing rate,moving window average,kurtosis and power spectral entropy [1]. So in total we extracted 31(channels) X 5 or 155 features for EEG signals.The EEG features were extracted at a sampling frequency of 100Hz for each EEG channel.…”

Section: Eeg and Speech Feature Extraction Detailsmentioning

confidence: 99%

“…We reduced the 155 EEG features to a dimension of 30 by applying Kernel Principle Component Analysis (KPCA) [20].We plotted cumulative explained variance versus number of components to identify the right feature dimension as shown in Figure 2. We used KPCA with polynomial kernel of degree 3 [1]. We further computed delta, delta and delta of those 30 EEG features, thus the final feature dimension of EEG was 90 (30 times 3) for both the data sets.…”

Section: Eeg and Speech Feature Extraction Detailsmentioning

confidence: 99%

“…Electroencephalography (EEG) is a non invasive way of measuring electrical activity of human brain. In [1] we demonstrated deep learning based automatic speech recognition (ASR) using EEG signals for a limited English vocabulary of four words and five vowels. In this paper we extend our work for a much larger English vocabulary and we use state-of-art end-to-end continuous speech recognition models to perform recognition.…”

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

Advancing Speech Recognition With No Speech Or With Noisy Speech

Krishna

Tran

Carnahan

et al. 2019

2019 27th European Signal Processing Conference (EUSIPCO)

Self Cite

View full text Add to dashboard Cite

show abstract

Section: Resultsmentioning

confidence: 86%

Section: Eeg and Speech Feature Extraction Detailsmentioning

confidence: 99%

Section: Eeg and Speech Feature Extraction Detailsmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Advancing Speech Recognition With No Speech Or With Noisy Speech

Krishna

Tran

Carnahan

et al. 2019

2019 27th European Signal Processing Conference (EUSIPCO)

Self Cite

View full text Add to dashboard Cite

show abstract

“…In [23] authors demonstrate that electroencephalography can be used for better performance of ASR system in the presence of noise. They used distillation training and showed that this training can improve the accuracy of an ASR system with EEG features in background noise.…”

Section: Review Of Literaturementioning

confidence: 99%

Speech Recognition Implementation

2019

IJITEE

View full text Add to dashboard Cite

This paper presents a brief review on Automatic Speech Recognition and provide a technical understanding of ASR system. The objective of this review paper is to elaborate one of the best techniques in the field of speech recognition that is hidden Markov model. Hidden Markov model is very popular technique for speech recognition because speech signal is more like piecewise stationary or short time stationary signal and these models can be trained easily and they are computationally feasible. So, this paper gives a proper implementation of hidden Markov model. After so many years of research, the main challenge in speech recognition field is accuracy. The speech recognition system includes feature extraction, building word template, comparing word and selecting the best with maximum likelihood. Hence, this paper will give a great contribution for understanding the concepts of Automatic Speech Recognition system and hidden Markov model.

show abstract

Leveraging AI‐enabled 6G‐driven IoT for sustainable smart cities

Gera,

Raghuvanshi,

Rawlley

et al. 2023

Int J Communication

View full text Add to dashboard Cite

SummaryMany scholastic researches have begun around the globe about the competitive technological interventions like 5G communication networks and its challenges. The incipient technology of 6G networks has emerged to facilitate ultrareliable and low‐latency applications for sustainable smart cities which are infeasible with the existing 4G/5G standards. Therefore, the advanced technologies like machine learning (ML), block chain, and Internet of Things (IoT) utilizing 6G network are leveraged to develop cost‐efficient mechanisms to address the issues of excess communication overhead in the present state of the art. Initially, the authors discussed the key vision of 6G communication technologies, its core technologies (such as visible light communication [VLC] and THz), and the existing issues with the existing network generations (such as 5G and 4G). A detailed analysis of benefits, challenges, and applications of blockchain‐enabled IoT devices with application verticals like Smart city, smart factory plus, automation, and XR that form the key highlights for 6G wireless communication network is also presented. In addition, the key applications and latest research of artificial intelligence (AI) in 6G are discussed facilitating the dynamic spectrum allocation mechanism and mobile edge computing. Lastly, an in‐depth study of the existing open issues and challenges in green 6G communication network technology, as well as review of solutions and potential research recommendations are also presented.

show abstract

Speech Recognition with No Speech or with Noisy Speech

Cited by 56 publications

References 30 publications

Advancing Speech Recognition With No Speech Or With Noisy Speech

Advancing Speech Recognition With No Speech Or With Noisy Speech

Speech Recognition Implementation

Leveraging AI‐enabled 6G‐driven IoT for sustainable smart cities

Contact Info

Product

Resources

About