Albayzín-2014 evaluation: audio segmentation and classification in broadcast news domains

@article{Castn2015Albayzn2014EA,
  title={Albayz{\'i}n-2014 evaluation: audio segmentation and classification in broadcast news domains},
  author={Diego Cast{\'a}n and David Tavarez and Paula Lopez-Otero and Javier Franco-Pedroso and H{\'e}ctor Delgado and Eva Navas and Laura Doc{\'i}o Fern{\'a}ndez and Daniel Ramos-Castro and Javier Serrano and Alfonso Ortega and Eduardo Lleida},
  journal={EURASIP Journal on Audio, Speech, and Music Processing},
  year={2015},
  volume={2015},
  pages={1-9}
}
Audio segmentation is important as a pre-processing task to improve the performance of many speech technology tasks and, therefore, it has an undoubted research interest. This paper describes the database, the metric, the systems and the results for the Albayzín-2014 audio segmentation campaign. In contrast to previous evaluations where the task was the segmentation of non-overlapping classes, Albayzín-2014 evaluation proposes the delimitation of the presence of speech, music and/or noise that… 

Figures and Tables from this paper

Audio Segmentation and Classification Approach Based on Adaptive CNN in Broadcast Domain
TLDR
This work proposes an adaptive CNN method which classifies clips directly based on audio sample points by increasing the amount of computation to ensure that the most suitable features can be extracted and achieves a significant error reduction.
Content-based audio classification and retrieval: A novel approach
  • Nilesh M. Patil, M. Nemade
  • Computer Science
    2016 International Conference on Global Trends in Signal Processing, Information Computing and Communication (ICGTSPICC)
  • 2016
TLDR
This paper gives the overview of various techniques used for classification and retrieval of audio and also proposes a novel approach for classified and retrieved of audio signal.
An Improved Approach to Audio Segmentation and Classification in Broadcasting Industries
TLDR
The segmentation-by-classification approach presents a novel novel smoothing-smoothing method(SEG-smoothness) and also presents the results of this method’s “segmentation- by- classification” experiments.
ALBAYZIN 2018 spoken term detection evaluation: a multi-domain international evaluation in Spanish
TLDR
The obtained results suggest that the STD task is still in progress and performance is highly sensitive to changes in the data domain.
ALBAYZIN 2016 spoken term detection evaluation: an international open competitive evaluation in Spanish
TLDR
An international open evaluation for search-on-speech based on STD in Spanish and an analysis of the results is presented.
Unsupervised Segmentation of Songs in Full Concert Audio
TLDR
This thesis proposes a solution to the unsupervised song segmentation in full concert audio, and is the first attempt to deal with the problem of this particular definition.
Music detection from broadcast contents using convolutional neural networks with a Mel-scale kernel
TLDR
The proposed method consistently showed better performance in all the three languages than the baseline system, and the F-score ranged from 86.5% for British data to 95.9% for Korean drama data.
Content-Based Audio Classification and Retrieval Using Segmentation, Feature Extraction and Neural Network Approach
TLDR
This paper proposes a machine learning- and neural network-based approach which performs audio pre-processing, segmentation, feature extraction, classification and retrieval of audio signal from the dataset using FPNN by combining fuzzy logic and PNN characteristics.
An Efficient Approach for Segmentation, Feature Extraction and Classification of Audio Signals
TLDR
The proposed approach achieves better performance in terms of precision, Normalized Mutual Information (NMI), F-score and entropy, and shows high False Rejection Rate (FRR), False Acceptance Rate (FAR), Genuine Acceptance rate (GAR) and accuracy with respect to the number of classes.
...
1
2
...

References

SHOWING 1-10 OF 42 REFERENCES
Audio segmentation of broadcast news in the Albayzin-2010 evaluation: overview, results, and discussion
TLDR
The evaluation of broadcast news audio segmentation systems carried out in the context of the Albayzín-2010 evaluation campaign is presented, with the aim of gaining an insight into the proposed solutions, and looking for directions which are promising.
Albayzin-2010 audio segmentation evaluation: evaluation setup and results
TLDR
The final results show that the problem of segmenting broadcast news is still challenging and the segmentation errors of the submitted systems are still challenging.
Audio segmentation-by-classification approach based on factor analysis in broadcast news domain
This paper studies a novel audio segmentation-by-classification approach based on factor analysis. The proposed technique compensates the within-class variability by using class-dependent factor
Broadcast news segmentation by audio type analysis
  • T. Nwe, Haizhou Li
  • Computer Science
    Proceedings. (ICASSP '05). IEEE International Conference on Acoustics, Speech, and Signal Processing, 2005.
  • 2005
TLDR
By analyzing the spectral properties of sound types, a multi-model HMM is proposed to integrate the primitive spectral properties in statistical modeling and it is shown that the proposed approach outperforms conventional methods.
Advances in unsupervised audio classification and segmentation for the broadcast news and NGSW corpora
TLDR
A new algorithm is proposed for audio classification, which is based on weighted GMM Networks (WGN), and a new false alarm compensation procedure is implemented, which can compensate the false alarm rate significantly with little cost to the miss rate.
Towards robust features for classifying audio in the CueVideo system
TLDR
This study modify and combine audio features known to be effective in distinguishing speech from music, and examine their behavior on mixed audio in the CueVideo system.
Content analysis for audio classification and segmentation
TLDR
A robust approach that is capable of classifying and segmenting an audio stream into speech, music, environment sound, and silence is proposed, and an unsupervised speaker segmentation algorithm using a novel scheme based on quasi-GMM and LSP correlation analysis is developed.
The COST278 broadcast news segmentation and speaker clustering evaluation - overview, methodology, systems, results
This paper describes a large scale experiment in which eight research institutions have tested their audio partitioning and labeling algorithms on the same data, a multi-lingual database of news
Classification of general audio data for content-based retrieval
Discrimination of speech from nonspeeech in broadcast news based on modulation frequency features
...
1
2
3
4
5
...