Automatic identification of sound recordings

Venkatachalam, V.; Cazzanti, Luca; Dhillon, Naveen; Wells, M G

doi:10.1109/msp.2004.1276117

Cited by 29 publications

(15 citation statements)

References 4 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…These papers do not address the response to compression. MusicDNA uses global mean and standard deviation of the energies within 15 subbands of 15 s of music, thus creating a 30-dimensional vector [5]. The effect of moderate compression is shown to be minimal.…”

Section: A Systems That Use Features Based On a Single Bandmentioning

confidence: 99%

Distortion Estimation in Compressed Music Using Only Audio Fingerprints

Doets

Lagendijk

2008

IEEE Trans. Audio Speech Lang. Process.

View full text Add to dashboard Cite

Abstract-An audio fingerprint is a compact yet very robust representation of the perceptually relevant parts of an audio signal. It can be used for content-based audio identification, even when the audio is severely distorted. Audio compression changes the fingerprint slightly. We show that these small fingerprint differences due to compression can be used to estimate the signal-to-noise ratio (SNR) of the compressed audio file compared to the original. This is a useful content-based distortion estimate, when the original, uncompressed audio file is unavailable. The method uses the audio fingerprints only. For stochastic signals distorted by additive noise, an analytical expression is obtained for the average fingerprint difference as function of the SNR level. This model is based on an analysis of the Philips robust hash (PRH) algorithm. We show that for uncorrelated signals, the bit error rate (BER) is approximately inversely proportional to the square root of the SNR of the signal. This model is extended to correlated signals and music. For an experimental verification of our proposed model, we divide the field of audio fingerprinting algorithms into three categories. From each category, we select an algorithm that is representative for that category. Experiments show that the behavior predicted by the stochastic model for the PRH also holds for the two other algorithms.

show abstract

Section: A Systems That Use Features Based On a Single Bandmentioning

confidence: 99%

Distortion Estimation in Compressed Music Using Only Audio Fingerprints

Doets

Lagendijk

2008

IEEE Trans. Audio Speech Lang. Process.

View full text Add to dashboard Cite

show abstract

“…The audio fingerprinting technique can be used in many applications [1], such as file sharing services, broadcast monitoring and so on. In digital rights management (DRM) system [2], the fingerprinting technique is also urgently required for the protection of Intelligent Property of the owner of media rights.…”

Section: Introductionmentioning

confidence: 99%

“…In [1], Haitsma and Kalker calculate the energy difference of the inter-frame and intra-frame and convert it to bit value and then use a sequence of bits to form an audio fingerprint. In [2], the square root of the mean energy across the time concatenating the standard deviation of the RMS power is used to form a fingerprint. The MPEG-7 audio descriptors-Audio Spectrum Flatness and Audio Signature are used to form the fingerprint in [3].…”

Section: Introductionmentioning

confidence: 99%

A Robust Feature Extraction Algorithm for Audio Fingerprinting

Chen

Huang

2008

Advances in Multimedia Information Processing - PCM 2008

View full text Add to dashboard Cite

Abstract. In this paper, we present a new feature extraction algorithm which can generate robust and reliable feature in a fingerprint system. This algorithm is referred to as weighted ASF (WASF). The feature in our algorithm is extracted based on a MPEG-7 descriptor-Audio Spectrum Flatness (ASF) and Human Auditory System (HAS). It also applies several effective filters to improve the feature robustness and uses another MPEG-7 descriptor: Audio Signature (AS) to reduce the feature dimension and increase the feature compactness. The smooth filter bank can efficiently resist the noise distortion in addition to some other common distortions such as sampling rate change and amplitude normalization, while the first order inverse filter can effectively resist the speed-change distortion with 90.1% discrimination for the 5% speed acceleration distortion. This algorithm is tested under several audio distortions: sampling rate change, noise addition, data compression and speedchange and so on. For these distortions, the WASF algorithm can get discrimination more than 90%. The MFCC feature and another MPEG-7 descriptor-Audio spectrum Centroid (ASC) are also considered.

show abstract

“…Audio watermarking finds applications in various areas such as copyright protection, data authentication, covert communications, addition of metadata, content identification and captioning or labeling of data [7,11,20]. In copyright application, the embedded data is intended to be used as a proof of digital media ownership to resolve copyright disputes.…”

Section: Introductionmentioning

confidence: 99%

An SVD-based audio watermarking technique

Özer

Sankur

Memon

2005

Proceedings of the 7th Workshop on Multimedia and Security

View full text Add to dashboard Cite

We present a non-oblivious, extremely robust watermarking scheme for audio signals. The watermarking algorithm is based on the SVD of the spectrogram of the signal. The SVD of the spectrogram is modified adaptively according to the information to be watermarked. The algorithm is tested for inaudibility performance with audio quality measures and robustness tests with audio Stirmark benchmark tool, which have a variety of common signal processing distortions. The comparison with a DCT based non-oblivious based method shows that the proposed SVD based method performs very satisfactorily.

show abstract

Automatic identification of sound recordings

Cited by 29 publications

References 4 publications

Distortion Estimation in Compressed Music Using Only Audio Fingerprints

Distortion Estimation in Compressed Music Using Only Audio Fingerprints

A Robust Feature Extraction Algorithm for Audio Fingerprinting

An SVD-based audio watermarking technique

Contact Info

Product

Resources

About