Mixture linear prediction Gammatone Cepstral features for robust speaker verification under transmission channel noise

Krobba, Ahmed; Debyeche, Mohamed; Selouani, Sid‐Ahmed

doi:10.1007/s11042-020-08748-2

Cited by 10 publications

(3 citation statements)

References 25 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The Gammatone frequency filter banks are indeed a group of cochlear modeling filter banks [39,40]. A Gammatone filter bank frequency selection is very close to the characteristics of an average human ear filter.…”

Section: Enhanced Gamma-tone Filter Bank Implementationmentioning

confidence: 96%

COVID-19 disease diagnosis with light-weight CNN using modified MFCC and enhanced GFCC from human respiratory sounds

Kumar

Alphonse

2022

Eur. Phys. J. Spec. Top.

View full text Add to dashboard Cite

In the last 2 years, medical researchers and clinical scientists have paid close attention to the problem of respiratory sound classification to classify COVID-19 disease symptoms. In the physical world, very few AI-based (Artificial Intelligence) techniques are often used to detect COVID-19/SARS-CoV-2 respiratory disease symptoms from the human respiratory system-generated acoustic sounds such as acoustic voice sound, breathing (inhale and exhale) sounds, and cough sound. We propose a light-weight Convolutional Neural Network (CNN) with Modified-Mel-frequency Cepstral Coefficient (M-MFCC) using different depths and kernel sizes to classify COVID-19 and other respiratory sound disease symptoms such as Asthma, Pertussis, and Bronchitis. The proposed network outperforms conventional feature extraction models and existing Deep Learning (DL) models for COVID-19/SARS-CoV-2 classification accuracy in the range of 4–10%. The model’s performance is compared with the COVID-19 crowdsourced benchmark dataset and gives a competitive performance. We applied different receptive fields and depths in the proposed model to get different contextual information that should aid in classification. And our experiments suggested 1 12 receptive fields and a depth of 5-Layer for the light-weight CNN to extract and identify the features from respiratory sound data. The model is also trained and tested with different modalities of data to showcase its effectiveness in classification.

show abstract

Section: Enhanced Gamma-tone Filter Bank Implementationmentioning

confidence: 96%

COVID-19 disease diagnosis with light-weight CNN using modified MFCC and enhanced GFCC from human respiratory sounds

Kumar

Alphonse

2022

Eur. Phys. J. Spec. Top.

View full text Add to dashboard Cite

show abstract

“…Dance features can be divided into continuous speech features, sound quality features, spectralbased features and non-linear teager energy operator (TEO)based features. Common spectral features include Linear Prediction Cepstral Coefficient (LPCC) (Krobba et al, 2020), Mel Frequency Cepstral Coefficient (MFCC) (Albadr et al, 2021), Log Frequency Power Coefficient (LFPC) (Gao, 2022). LPCC linearly approximates speech at all frequencies, which is inconsistent with human auditory characteristics.…”

Section: Features Of Dance Emotionmentioning

confidence: 99%

Dance emotion recognition based on linear predictive Meir frequency cepstrum coefficient and bidirectional long short-term memory from robot environment

Shen

Xiao-xi²,

Jiang³

et al. 2022

Front. Neurorobot.

View full text Add to dashboard Cite

Dance emotion recognition is an important research direction of automatic speech recognition, especially in the robot environment. It is an important research content of dance emotion recognition to extract the features that best represent speech emotion and to construct an acoustic model with strong robustness and generalization. The dance emotion data set is small in size and high in dimension. The traditional recurrent neural network (RNN) has the problem of long-range dependence disappearance, and due to the focus on local information of convolutional neural network (CNN), the mining of potential relationships between frames in the input sequence is insufficient. To solve the above problems, this paper proposes a novel linear predictive Meir frequency cepstrum coefficient and bidirectional long short-term memory (LSTM) for dance emotion recognition. In this paper, the linear prediction coefficient (LPC) and Meier frequency cepstrum coefficient (MFCC) are combined to obtain a new feature, namely the linear prediction Meier frequency cepstrum coefficient (LPMFCC). Then, the combined feature obtained by combining LPMFCC with energy feature is used as the extracted dance feature. The extracted features are input into the bidirectional LSTM network for training. Finally, support vector machine (SVM) is used to classify the obtained features through the full connection layer. Finally, we conduct experiments on public data sets and obtain the better effectiveness compared with the state-of-art dance motion recognition methods.

show abstract

“…Most feature extraction methods use the Karhunen-Loeve Transform (KLT) [9,10]. With outstanding results, these methods were applied to cases of text-independent speaker recognition [8].…”

Section: Introductionmentioning

confidence: 99%

Text-independent speaker identification system using discrete wavelet transform with linear prediction coding

Alrusaini,

Daqrouq

2024

J. Umm Al-Qura Univ. Eng.Archit.

View full text Add to dashboard Cite

One of the key problems of the modern day is the presentation of an identity verification system that can perform sufficient accuracy in identity verification, is resilient to assaults and noises, and can be recorded in the simplest possible method. In this study, a new speaker feature extraction which based on discrete wavelet transform (DWT) and linear prediction coding (LPC) algorithm (WLPCA) are investigated. This paper's primary objective is to evidence the performance of the new method for speaker identification by a Gaussian mixture model (GMM). The proposed method improves the recognition rate over the Mel-frequency cepstral coefficient (MFCC). Experimental evaluation of the process performance is performed on two speech databases; our recorded database and the publicly available TIMIT database. We show that the speech features derived by the newly proposed method are more suitable for GMM (91.53%), in terms of the time-consumed, by requiring less Gaussian mixtures than MFCC (85.77%). For testing the presented method in a noisy environment, Additive white Gaussian noise (AWGN) was added to the TIMIT database, where a slight improvement in the performance of the presented method (60.02%) over the MFCC (59.89%) was observed.

show abstract

Mixture linear prediction Gammatone Cepstral features for robust speaker verification under transmission channel noise

Cited by 10 publications

References 25 publications

COVID-19 disease diagnosis with light-weight CNN using modified MFCC and enhanced GFCC from human respiratory sounds

COVID-19 disease diagnosis with light-weight CNN using modified MFCC and enhanced GFCC from human respiratory sounds

Dance emotion recognition based on linear predictive Meir frequency cepstrum coefficient and bidirectional long short-term memory from robot environment

Text-independent speaker identification system using discrete wavelet transform with linear prediction coding

Contact Info

Product

Resources

About