The Linear Predictive Modeling of Speech From Higher-Lag Autocorrelation Coefficients Applied to Noise-Robust Speaker Recognition

Alku, Paavo; Saeidi, Rahim

doi:10.1109/taslp.2017.2703165

Cited by 13 publications

(4 citation statements)

References 49 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The Kannada speech signal is decomposed into a number of frames by multipling window function of length L using (1). Then each frame average short time energy [10][11] is computed using (2).…”

Section: Average Short Time Energy (Ste)mentioning

confidence: 99%

See 1 more Smart Citation

Continuous kannada speech segmentation and speech recognition based on threshold using MFCC And VQ

Gowda¹,

Mathivanan²,

Thangamuthu³

2019

IJECE

View full text Add to dashboard Cite

<p><span>Continuous speech segmentation and its recognition is playing important role in natural language processing. Continuous context based Kannada speech segmentation depends on context, grammer and semantics rules present in the kannada language. The significant feature extraction of kannada speech signal for recognition system is quite exciting for researchers. In this paper proposed method is divided into two parts. First part of the method is continuous kannada speech signal segmentation with respect to the context based is carried out by computing average short term energy and its spectral centroid coefficients of the speech signal present in the specified window. The segmented outputs are completely meaningful segmentation for different scenarios with less segmentation error. The second part of the method is speech recognition by extracting less number Mel frequency cepstral coefficients with less number of codebooks using vector quantization .In this recognition is completely based on threshold value.This threshold setting is a challenging task however the simple method is used to achieve better recognition rate.The experimental results shows more efficient and effective segmentation with high recognition rate for any continuous context based kannada speech signal with different accents for male and female than the existing methods and also used minimal feature dimensions for training data.</span></p>

show abstract

Section: Average Short Time Energy (Ste)mentioning

confidence: 99%

“…Higher lag is method is used to extract the features of speech signal with linear prediction [1]. The method gives two prediction errors, one is the ordinary convention linear prediction and other one is the delayed version with k number of samples of linear prediction.…”

Section: Introductionmentioning

confidence: 99%

Continuous kannada speech segmentation and speech recognition based on threshold using MFCC And VQ

Gowda¹,

Mathivanan²,

Thangamuthu³

2019

IJECE

View full text Add to dashboard Cite

show abstract

“…Linear predictive analysis of the speech signal is one of the most powerful speech analysis techniques, which can extract the short-time spectral envelope information of speech signals efficiently, and is widely used in the fields of speech representing for low bit rate transmission or storage, automatic speech and speaker recognition [1][2][3][4][5]. The predominant linear predictive analysis method is Linear Prediction Coding (LPC), featuring better fault tolerance during transmitting spectral envelope information.…”

Section: Introductionmentioning

confidence: 99%

Fast Computation of LSP Frequencies Using the Bairstow Method

et al. 2020

View full text Add to dashboard Cite

Linear prediction is the kernel technology in speech processing. It has been widely applied in speech recognition, synthesis, and coding, and can efficiently and correctly represent the speech frequency spectrum with only a few parameters. Line Spectrum Pairs (LSPs) frequencies, as an alternative representation of Linear Predictive Coding (LPC), have the advantages of good quantization accuracy and low spectral sensitivity. However, computing the LSPs frequencies takes a long time. To address this issue, a fast computation algorithm, based on the Bairstow method for computing LSPs frequencies from linear prediction coefficients, is proposed in this paper. The algorithm process first transforms the symmetric and antisymmetric polynomial to general polynomial, then extracts the polynomial roots. Associated with the short-term stationary property of speech signal, an adaptive initial method is applied to reduce the average iteration numbers by 26%, as compared to the statics in the initial method, with a Perceptual Evaluation of Speech Quality (PESQ) score reaching 3.46. Experimental results show that the proposed method can extract the polynomial roots efficiently and accurately with significantly reduced computation complexity. Compared to previous works, the proposed method is 17 times faster than Tschirnhus Transform, and has a 22% PESQ improvement on the Birge-Vieta method with an almost comparable computation time.

show abstract

“…Multi-channel Linear Prediction (LP) is widely used for signal processing techniques and applications such as signal compression, beamforming [11] speech enhancement [12,13] and speech recognition [14]. In the context of ad-hoc signal processing it is important to modify the standard multi-channel LP in order to consider the wide spatial coverage of the array as well as inconsistencies in microphone gains.…”

Section: Introductionmentioning

confidence: 99%

Multi-Channel Compression and Coding of Reverberant Ad-Hoc Recordings Through Spatial Autoregressive Modelling

Pasha

Lundgren

2019

2019 30th Irish Signals and Systems Conference (ISSC)

View full text Add to dashboard Cite

Autoregressive modelling techniques such as multi-channel linear prediction are widely used for applications such as coding, dereverberation and compression of the speech signals. State of the art multi-channel linear prediction methods do not take into account the spatial locations of the microphones and assume single distance compact microphone arrays. In this paper a spatially modified multichannel autoregressive compression and coding method is proposed and successfully tested in order to adapt the standard multi-channel method to the virtual reality and immersive video conferencing applications where the microphones can be meters away from each other. The proposed method estimates the spatial distances between each microphone and the source to optimise the joint compression of the microphones covering a wide area. The results suggest that the proposed method outperforms the standard multi-channel compression and coding method when applied to the ad-hoc scenarios.

show abstract

The Linear Predictive Modeling of Speech From Higher-Lag Autocorrelation Coefficients Applied to Noise-Robust Speaker Recognition

Cited by 13 publications

References 49 publications

Continuous kannada speech segmentation and speech recognition based on threshold using MFCC And VQ

Continuous kannada speech segmentation and speech recognition based on threshold using MFCC And VQ

Fast Computation of LSP Frequencies Using the Bairstow Method

Multi-Channel Compression and Coding of Reverberant Ad-Hoc Recordings Through Spatial Autoregressive Modelling

Contact Info

Product

Resources

About