2019 7th International Conference on Smart Computing &Amp; Communications (ICSCC) 2019
DOI: 10.1109/icscc.2019.8843660
|View full text |Cite
|
Sign up to set email alerts
|

DWT-MFCC Method for Speaker Recognition System with Noise

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
3
1

Citation Types

0
15
0

Year Published

2020
2020
2024
2024

Publication Types

Select...
4
3

Relationship

0
7

Authors

Journals

citations
Cited by 10 publications
(15 citation statements)
references
References 18 publications
0
15
0
Order By: Relevance
“…Records are separated by finite impulse response (FIR) [25], [26]. The result shown in Figure 3(b) for channel 1 shows that the noise tends to decrease compared to channel 2 in Figure 3(c The word samples that have been separation are then extracted using MFCC with the following steps; frame blocking, windowing, fast Fourier transform (FFT), mel scale filterbank, discrete cosine transform (DCT) and mel frequency cepstral coefficients [12]. The frequency component contained in the sample words in each channel, the frequency value is obtained by dividing the sample words into several frames.…”
Section: Mfcc Dual-channelmentioning
confidence: 99%
See 2 more Smart Citations
“…Records are separated by finite impulse response (FIR) [25], [26]. The result shown in Figure 3(b) for channel 1 shows that the noise tends to decrease compared to channel 2 in Figure 3(c The word samples that have been separation are then extracted using MFCC with the following steps; frame blocking, windowing, fast Fourier transform (FFT), mel scale filterbank, discrete cosine transform (DCT) and mel frequency cepstral coefficients [12]. The frequency component contained in the sample words in each channel, the frequency value is obtained by dividing the sample words into several frames.…”
Section: Mfcc Dual-channelmentioning
confidence: 99%
“…The process is by matching the features of evidence and comparison. Similarity of features means that the audio records come from the same individual [12]. The similarity of features is calculated from the high value of accuracy.…”
Section: Introductionmentioning
confidence: 99%
See 1 more Smart Citation
“…It is stated that the features extracted using a wavelet give equal accuracy to MFCC. Aside from that, many previous kinds of research claim that extracting features using wavelet yielded better results compared to MFCC [9]. WPT was employed to extract features on TIMIT and TALUNG databases [10].…”
Section: Introductionmentioning
confidence: 99%
“…MFCC can represent the speech spectrum in a compact form and is based on the model of human auditory perception, but https://doi.org/10.31436/iiumej.v23i1.1760 the FFT process can cause the loss of information [5,11]. Meanwhile, the wavelet can map the signal into the time and frequency domain; thus, no information is lost in this process [9]. MFCC was combined with 3 level decomposition of DWT to extract signal uttering digits 0-9 in the Indian language [12].…”
Section: Introductionmentioning
confidence: 99%