AMADEUS: a scalable HMM-based audio information retrieval system

Batlle, Eloi; Masip, Jaume; Guaus, Enric

doi:10.1109/isccsp.2004.1296517

Cited by 3 publications

(5 citation statements)

References 8 publications

(2 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…energy-based features, or auto correlation values). In [1,2], for example, MFCCs are used for HMMbased song identification whereas in [9] statistical features describing pitch, rhythm etc. are applied for mixture-density based genre classification.…”

Section: Related Workmentioning

confidence: 99%

See 1 more Smart Citation

Automatic Detection of Song Changes in Music Mixes Using Stochastic Models

Plötz

Fink

Husemann

et al. 2006

18th International Conference on Pattern Recognition (ICPR'06)

View full text Add to dashboard Cite

The annotation of song changes in music mixes created by DJs or radio stations for direct access in digital recordings is, usually, a very tedious work. In order to support this process we developed an automatic song change detection method which can be used for arbitrary music mixes. Stochastic models are applied to music data aiming at their segmentation with respect to automatically obtained abstract generic acoustic units. The local analysis of these stochastic music models provides hypotheses for song changes. Results of an experimental evaluation processing music mix data demonstrate the effectiveness of our method for supporting the annotation with respect to song changes.

show abstract

Section: Related Workmentioning

confidence: 99%

“…In [1,2] a system for song identification in radio broadcasts has been presented. The authors use an MFCC-based feature representation of music for un-supervised training of HMMs for modeling generic acoustic generators (GAGs).…”

Section: Related Workmentioning

confidence: 99%

Automatic Detection of Song Changes in Music Mixes Using Stochastic Models

Plötz

Fink

Husemann

et al. 2006

18th International Conference on Pattern Recognition (ICPR'06)

View full text Add to dashboard Cite

show abstract

“…Many content-based audio information retrieval (AIR) systems have been proposed to date [3][4][5][6][7]. They typically extract a set of acoustic features such as pitch, duration and rhythm from audio and transcribe them into some symbolic representation for efficient audio retrieval.…”

Section: Introductionmentioning

confidence: 99%

M-MUSICS: an intelligent mobile music retrieval system

2010

View full text Add to dashboard Cite

Accurate voice humming transcription and efficient indexing and retrieval schemes are essential to a large-scale humming-based audio retrieval system. Although much research has been done to develop such schemes, their performance in terms of precision, recall, and F-measure, among all similarity metrics, are still unsatisfactory. In this paper, we propose a new voice query transcription scheme. It considers the following features: note onset detection using dynamic threshold methods, fundamental frequency (F0) acquisition of each frame, and frequency realignment using K-means. We use a popularity-adaptive indexing structure called frequently accessed index (FAI) based on frequently queried tunes for indexing purposes. In addition, we propose a semi-supervised relevance feedback and query reformulation scheme based on a genetic algorithm to improve retrieval efficiency. In this paper, we extend our efforts to mobile multimedia environments and develop a mobile audio retrieval system. Experiments show our system performs satisfactory in wireless mobile multimedia environments.

show abstract

“…All the probabilities derived can be seen as a distance between two songs and, therefore, a metric (with some restrictions) can be defined with acoustical similarities as explained in [2]. …”

Section: Rationalementioning

confidence: 99%

“…All the training process is based on a modification of the Baum-Welch algorithm (also based o n the EM algorithm [4]) and is explained in further detail in [2].…”

Section: Training Of the Systemmentioning

confidence: 99%

Scalability issues in an HMM-based audio fingerprinting

Batlle

Masip

Guaus

et al.

2004 IEEE International Conference on Multimedia and Expo (ICME) (IEEE Cat. No.04TH8763)

Self Cite

View full text Add to dashboard Cite

Audio fingerprinting technologies allow the identification of audio content without the need of an extemal metadata or watermark embedding. These audio fingerprinting technologies work by extracting a content-based compact digest that summarizes a recording and comparing them wifh a previously extractedfingerprint database. In this paper: we present afingerprint scheme that is based on Hidden Markov Models. This approach achieves a high compaction of the audio signal by exploiting structural redundancies on music and robustness to distorfions thanks to the stochastic modeling. In this paper we present the basic functionality of the system as well as some results.

show abstract

AMADEUS: a scalable HMM-based audio information retrieval system

Cited by 3 publications

References 8 publications

Automatic Detection of Song Changes in Music Mixes Using Stochastic Models

Automatic Detection of Song Changes in Music Mixes Using Stochastic Models

M-MUSICS: an intelligent mobile music retrieval system

Scalability issues in an HMM-based audio fingerprinting

Contact Info

Product

Resources

About