Audio Bank: A high-level acoustic signal representation for audio event recognition

Sandhan, Tushar; Sonowal, Sukanya; Choi, Jin Young

doi:10.1109/iccas.2014.6987963

Cited by 8 publications

(13 citation statements)

References 16 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…As is known to all, many previous researches on audio classification have been done to prove the effectiveness of Hidden Markov Model (HMM) [6,9]. In this paper, we utilize HMM to train the signal audio effects.…”

Section: Basic Audio Element Modelingmentioning

confidence: 99%

“…The SVMbased method is introduced in [5], and the Gaussian radial basis function is used as the kernel function. The HMMbased method is introduced in [6], which has been widely used in creating keywords thus to be retrieved in movies. According to [6], the state number for each abnormal audio is set in Table 5.…”

Section: Evaluation Of the Background Pool Modelmentioning

confidence: 99%

“…The HMMbased method is introduced in [6], which has been widely used in creating keywords thus to be retrieved in movies. According to [6], the state number for each abnormal audio is set in Table 5. For each target abnormal audio event, we use the precision and recall to evaluate its detection accuracy:…”

Section: Evaluation Of the Background Pool Modelmentioning

confidence: 99%

“…We extract the MFCCs for each audio clip, and then the extracted MFCCs vectors are used as the input observations for the HMMs. According to some previous works [6], the Baum-Welch algorithm is then applied to estimate the transition probabilities between states and the observation probabilities in each state. After that, we have built the model for each basic audio element.…”

Section: Basic Audio Element Modelingmentioning

confidence: 99%

“…This system can detect some impulsive sound such as door alarm and crying. In [6], the authors designed a HMM-based method to detect some special audio elements such as gun-shot and car-crashing. However, in some special monitoring systems (e.g., in the forest monitoring system), there is no need to distinguish gun-shot from animal scream; it is necessary to judge whether the event is expected to happen at a specific time and a specific location.…”

Section: Introductionmentioning

confidence: 99%

See 4 more Smart Citations

Abnormal Event Detection Method in Multimedia Sensor Networks

Liu

Yang

et al. 2015

International Journal of Distributed Sensor Networks

View full text Add to dashboard Cite

Detecting abnormal events in multimedia sensor networks (MSNs) plays an increasingly essential role in our lives. Once video cameras cannot work (e.g., the sightline is blocked), audio sensor can provide us with critical information (e.g., in detecting the sound of gun-shot in the rainforest or the sound of car accident on a busy road). Audio sensors also have price advantage. Detecting abnormal audio events in complicated background environment is a very difficult problem; only few previous researches could offer good solution. In this paper, we proposed a novel method to detect the unexpected audio elements in multimedia sensor networks. Firstly, we collect enough normal audio elements and then use statistical learning method to train them offline. On the basis of these models, we establish a background pool by prior knowledge. The background pool contains expected audio effects. Finally, we decide whether an audio event is unexpected by comparing it with the background pool. In this way, we reduce the complexity of online training while ensuring the detection accuracy. We designed some experiments to verify the effectiveness of the proposed method. In conclusion, the experiments show that the proposed algorithm can achieve satisfying results.

show abstract

Section: Basic Audio Element Modelingmentioning

confidence: 99%