2005
DOI: 10.1155/asp.2005.1780
|View full text |Cite
|
Sign up to set email alerts
|

Perceptual Audio Hashing Functions

Abstract:

Perceptual hash functions provide a tool for fast and reliable identification of content. We present new audio hash functions based on summarization of the time-frequency spectral characteristics of an audio document. The proposed hash functions are based on the periodicity series of the fundamental frequency and on singular-value description of the cepstral frequencies. They are found, on one hand, to perform very satisfactorily in identification and verification tests, and on the other hand, to be ve… Show more

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
2
1
1
1

Citation Types

0
32
0

Year Published

2007
2007
2015
2015

Publication Types

Select...
5
2
1

Relationship

0
8

Authors

Journals

citations
Cited by 40 publications
(32 citation statements)
references
References 17 publications
0
32
0
Order By: Relevance
“…For speech signals, the first 13 cepstrum coefficients are often utilized [28]; hence, we considered first 13 MFCC features for the proposed CBCD task.…”
Section: Mfccs Extractionmentioning
confidence: 99%
“…For speech signals, the first 13 cepstrum coefficients are often utilized [28]; hence, we considered first 13 MFCC features for the proposed CBCD task.…”
Section: Mfccs Extractionmentioning
confidence: 99%
“…Given this, the proposed algorithm firstly performs the intensity-loudness transform (ILT) to the speech signal after preprocessing. The nonlinear relationships between loudness L and intensity I are shown as in (4).…”
Section: B Wavelet Decompositionmentioning
confidence: 99%
“…The features extracted are mainly that logarithmic cepstrum coefficient [4], linear spectrum frequencies [5], MFCCs [6] and LPCCs [7]. Other feature extraction methods are also widely adopted, such as Hilbert transform spectrum estimation [8], temporal modulation normalization [9] and bark-bands energy [10], etc.…”
Section: Introductionmentioning
confidence: 99%
“…This algorithm does not reflect the distortions related to compression, especially at medium and high bit rates. Özer et al use periodicity estimators and a singular value decomposition of the Mel frequency cepstrum coefficient (MFCC) matrix [3]. Sukkittanon and Atlas propose frequency modulation features [4].…”
Section: A Systems That Use Features Based On a Single Bandmentioning
confidence: 99%