Sentence examples for audio retrieval from inspiring English sources

The experiments of audio retrieval demonstrate that Gaussian-LDA achieves better performance than other compared methods.

In [11], scalability of several classification methods is analyzed for large-scale audio retrieval.

A content-based audio retrieval application was presented to explain the basic blocks of audio classification.

Figure 1 Operating modes of hybrid network for audio retrieval and annotation.

However, most of those methods are designed for multimedia retrieval in single modality, such as image retrieval and audio retrieval.

To overcome this shortage, this paper introduces a new topic model named Gaussian-LDA for audio retrieval.

A comparison of five similarity metrics from the WordNet::Similarity library in terms of audio information retrieval was studied in [15].

Finally, using a domain-specific ontology such as the MX music ontology [27] might be better suited to audio information retrieval than a purely lexical database such as WordNet.

This flexibility is an important advantage of the hybrid network approach as compared to the multiclass supervised leaning approaches to audio information retrieval, for example, [7, 9]. Figure 2 displays an example hybrid network illustrating the difference between in- and out-of-vocabulary semantic tags.

The presented work focuses on one such way for automatic classification of audio signals for retrieval purposes.

The main motivation for the work is to provide improved content-based audio classification and retrieval performance.