Useful in many applications, text mining techniques offer the possibility to analyze texts and discover their semantic meaning. Unfortunately, most of these methods only work on textual data, but they are equally useful when dealing with spoken conversations. In this contribution we show how an open source and freely available Automatic Speech Recognition engine can be used out-of-the-box as a preprocessing step for keyword spotting by first transcribing spoken conversations. Can lower quality transcriptions still prove useful for keyword extraction?
|Journal||Belgian/Netherlands Artificial Intelligence Conference|
|Publication status||Published - 2011|
|Event||23rd Benelux Conference on Artificial Intelligence, BNAIC 2011 - Ghent, Belgium|
Duration: 3 Nov. 2011 → 4 Nov. 2011