Speech synthesis and recognition are used in a wide variety of fields, such as voice assistants, IVR systems and smart homes.


Speech analysis and machine learning International business: competing in the global marketplace
Speech analysis and machine learning

Text analysis

The methods in fetching shorthand work on a subset selection basis. It does this by extracting phrases or sentences from an article to form a resume. LexRank and TextRank are well-known representatives of this approach, which use variations on the Google PageRank page sorting algorithm.


LexRank is a graph-based unsupervised learning algorithm that uses the modified cosine of the inverse frequency of occurrence of a word as a measure of the similarity of two sentences. The similarity is used as the weight of the graph edge between two sentences. LexRank is also implementing a smart post-processing step that makes sure the top sentences are not too similar to each other.