[HTML][HTML] Environmentally robust ASR front-end for deep neural network acoustic models
T Yoshioka, MJF Gales - Computer Speech & Language, 2015 - Elsevier
This paper examines the individual and combined impacts of various front-end approaches
on the performance of deep neural network (DNN) based speech recognition systems in …
on the performance of deep neural network (DNN) based speech recognition systems in …
Investigation of unsupervised adaptation of DNN acoustic models with filter bank input
Adaptation to speaker variations is an essential component of speech recognition systems.
One common approach to adapting deep neural network (DNN) acoustic models is to …
One common approach to adapting deep neural network (DNN) acoustic models is to …
Study of statistical robust closed set speaker identification with feature and score-based fusion
MTS Al-Kaltakchi, WL Woo, SS Dlay… - 2016 IEEE statistical …, 2016 - ieeexplore.ieee.org
In this paper, the statistical combination of Power Normalization Cepstral Coefficient (PNCC)
and Mel Frequency Cepstral Coefficient (MFCC) features in robust closed set speaker …
and Mel Frequency Cepstral Coefficient (MFCC) features in robust closed set speaker …
Spectro-temporal power spectrum features for noise robust ASR
In this paper, we present a new technique to extract a noise robust representation of speech
signals called spectro-temporal power spectrum. This technique is based on applying a …
signals called spectro-temporal power spectrum. This technique is based on applying a …
A compact CNN-based speech enhancement with adaptive filter design using gabor function and region-aware convolution
S Abdullah, M Zamani, A Demosthenous - IEEE Access, 2022 - ieeexplore.ieee.org
Speech enhancement (SE) is used in many applications, such as hearing devices, to
improve speech intelligibility and quality. Convolutional neural network-based (CNN-based) …
improve speech intelligibility and quality. Convolutional neural network-based (CNN-based) …
Kernel power flow orientation coefficients for noise-robust speech recognition
B Gerazov, Z Ivanovski - IEEE/ACM Transactions on Audio …, 2014 - ieeexplore.ieee.org
Noise-robustness has become a crucial parameter in Automatic Speech Recognition (ASR)
systems today with their increased use in noise-filled real-world environments. One way to …
systems today with their increased use in noise-filled real-world environments. One way to …
On the importance of modeling and robustness for deep neural network feature
A large body of research has shown that acoustic features for speech recognition can be
learned from data using neural networks with multiple hidden layers (DNNs) and that these …
learned from data using neural networks with multiple hidden layers (DNNs) and that these …
[PDF][PDF] Comparing time-frequency representations for directional derivative features.
We compare the performance of Directional Derivatives features for automatic speech
recognition when extracted from different time-frequency representations. Specifically, we …
recognition when extracted from different time-frequency representations. Specifically, we …
Robust Features in Deep Neural Networks for Transcoded Speech Recognition DSR and AMR-NB
L Bouchakour, M Debyeche… - 2024 8th International …, 2024 - ieeexplore.ieee.org
Automatic Speech Recognition (ASR) performance in mobile communications degrades
significantly if the environment includes many sources of variability, such as when the test …
significantly if the environment includes many sources of variability, such as when the test …
[PDF][PDF] Time-Frequency Coherence for Periodic-Aperiodic Decomposition of Speech Signals.
Decomposing speech signals into periodic and aperiodic components is an important task,
finding applications in speech synthesis, coding, denoising, etc. In this paper, we construct a …
finding applications in speech synthesis, coding, denoising, etc. In this paper, we construct a …