Broadcast news audio classification using SVM binary trees

Vavrek, Jozef; Vozarikova, Eva; Juhar, Jozef

doi:10.1109/tsp.2012.6256338

Cited by 13 publications

(5 citation statements)

References 13 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Temporal Audio Features: these features are directly calculated from the audio samples. The most known timedomain features are Short-term energy [98] [99], energy entropy and zero crossing rate [98]. These features will be utilized in the feature extraction stage of our technique because they guarantee simple and good mean for audio signals analysis.…”

Section: ) Proposed Scheme For Content Characterization Using Audio Watermarkingmentioning

confidence: 99%

“…These metrics are computed using Discrete Fourier Transform (DFT) coefficients of the designed audio frame. The most known spectral-domain features are spectral flux , spectral centroid, [100], spectral roll off, Mel-Frequency Cepstrum Coefficients (MFCCs) [98,100], chroma vector [101] and Relative Spectral Analysis-Perceptual Linear Prediction (Rasta PLP) [102].…”

Section: ) Proposed Scheme For Content Characterization Using Audio Watermarkingmentioning

confidence: 99%

See 1 more Smart Citation

Audio Watermarking for Security and Non-Security Applications

et al. 2022

View full text Add to dashboard Cite

The digitization of audiovisual data is significantly increasing. Thus, in order to guarantee principally the protection of intellectual properties of this digital content, watermarking has appeared as a solution. The watermarking can be used in reality in several types of applications that target two different contexts; the first for security applications and the second for non-security ones. In this paper, we carry a big interest in studying these two types of applications. Moreover, we propose a first digital watermarking scheme for security copyright protection application where we have involved Neural Network architecture in the insertion and detection processes and we have integrated some masking phenomena of the Human Psychoacoustic Model with Linear Predictive Coding spectral envelope estimation of the audio file. Experimentations proved the efficiency of exploiting perceptual masking with spectral envelope consideration in terms of imperceptibility and robustness results. Besides, we suggest a second audio watermarking technique for non-security content characterization application based on deep learning classification architecture. In this scheme, extracted watermark will advise about the audio class: music or speech, the speaker gender and emotion. Reported results indicated that the suggested scheme achieved higher performance at classification level as well as at watermarking properties.

show abstract

Section: ) Proposed Scheme For Content Characterization Using Audio Watermarkingmentioning

confidence: 99%

Section: ) Proposed Scheme For Content Characterization Using Audio Watermarkingmentioning

confidence: 99%

Audio Watermarking for Security and Non-Security Applications

et al. 2022

View full text Add to dashboard Cite

show abstract

“…Since the broadcasting FM channels demand the news content classification of broadcasting context, Vavrek, J. et al [20] also proposed a hierarchical tree to address the news content classification problem. This hierarchical classification strategy is used as a particular feature set for each SVM binary classifier.…”

Section: Same As Neural Network Approaches the Hidden Markovmentioning

confidence: 99%

“…In addition to that, 16000 Hz and 22050 Hz were also used as the sample rates in previous works related to radio broadcasting classification. Weeratunga et al [15] proposed 22050 Hz as the sample rate, John Saunders [16] and Vavrek, J. et al [20] proposed 16000 Hz as the sample rates for their studies. Therefore, we evaluate the models with sample rates 16000 Hz and 22050 Hz, and 44100 Hz to find the most reliable sample rate for the research.…”

Section: B Different Frame Sizesmentioning

confidence: 99%

Classification of Voice Content in the Context of Public Radio Broadcasting

Karunarathna

Jayaratne

Gunawardana

2020

Int J on Adv. in ICT for Emerging Countries

View full text Add to dashboard Cite

With the rapid development of mass media technology, content classification of radio broadcasting has emerged as a major research area facilitating the automation of radio broadcasting monitoring process. This research focuses on the voice dominant content classification of radio broadcasting by employing a multi-class Support Vector Machine (SVM) in order to automate monitoring of radio broadcasting in Sri Lanka. This study investigates the performance of "One Vs. One" and "One Vs. All" methods known to be two conventional ways to build a multi-class SVM. These two multi-class SVM models are trained to classify five voice dominant classes as news, conversations, and advertisements without jingles, radio drama and religious programs.One of the substantial measures in creating such a classification is selection of the optimal feature sets. For that purpose, time domain features, frequency domain features, cepstral features, and chroma features are manually analyzed for each binary SVM classifier independently. Two multi-class SVM models are trained based on the selected features and the "One Vs. One" model was able to better classify the recordings with an 85% accuracy compared to 83% accuracy achieved by "One Vs. All" model. Further, the results revealed the importance of careful feature selection in order to achieve higher classification accuracies.

show abstract

“…2. Methods based on the template matching of features extracted from the query/utterance speech signal [7,8,[17][18][19][20][21][22][23][24][25][26][27][28][29]. They usually borrow the idea from dynamic time warping (DTW)-based speech recognition and were found to outperform phonetic transcription-based techniques on QbE STD [18].…”

Section: Introductionmentioning

confidence: 99%

Query-by-Example Spoken Term Detection ALBAYZIN 2012 evaluation: overview, systems, results, and discussion

Tejedor

Toledano

Anguera

et al. 2013

J AUDIO SPEECH MUSIC PROC.

View full text Add to dashboard Cite

Query-by-Example Spoken Term Detection (QbE STD) aims at retrieving data from a speech data repository given an acoustic query containing the term of interest as input. Nowadays, it has been receiving much interest due to the high volume of information stored in audio or audiovisual format. QbE STD differs from automatic speech recognition (ASR) and keyword spotting (KWS)/spoken term detection (STD) since ASR is interested in all the terms/words that appear in the speech signal and KWS/STD relies on a textual transcription of the search term to retrieve the speech data. This paper presents the systems submitted to the ALBAYZIN 2012 QbE STD evaluation held as a part of ALBAYZIN 2012 evaluation campaign within the context of the IberSPEECH 2012 Conference a . The evaluation consists of retrieving the speech files that contain the input queries, indicating their start and end timestamps within the appropriate speech file. Evaluation is conducted on a Spanish spontaneous speech database containing a set of talks from MAVIR workshops b , which amount at about 7 h of speech in total. We present the database metric systems submitted along with all results and some discussion. Four different research groups took part in the evaluation. Evaluation results show the difficulty of this task and the limited performance indicates there is still a lot of room for improvement. The best result is achieved by a dynamic time warping-based search over Gaussian posteriorgrams/posterior phoneme probabilities. This paper also compares the systems aiming at establishing the best technique dealing with that difficult task and looking for defining promising directions for this relatively novel task.

show abstract

Broadcast news audio classification using SVM binary trees

Cited by 13 publications

References 13 publications

Audio Watermarking for Security and Non-Security Applications

Audio Watermarking for Security and Non-Security Applications

Classification of Voice Content in the Context of Public Radio Broadcasting

Query-by-Example Spoken Term Detection ALBAYZIN 2012 evaluation: overview, systems, results, and discussion

Contact Info

Product

Resources

About