Details

Autor(en) / Beteiligte

• Hamzah, Raseeda
• Jamil, Nursuriati
• Seman, Noraini
• Jaafar, Azizah
• Bakar, Zainab Abu
• Sembok, Tengku Mohd Tengku
• Smeaton, Alan F.
• Bruza, Peter
• Mohd Noah, Shahrul Azman
• Mohamad Ali, Nazlena

Titel

Nurturing Filled Pause Detection for Spontaneous Speech Retrieval

Ist Teil von

• Information Retrieval Technology, p.458-469

Ort / Verlag

Cham: Springer International Publishing

Link zum Volltext

Quelle

Alma/SFX Local Collection

Beschreibungen/Notizen

• In this paper we investigate methods to adapt a system for filled pause (FP) disfluency removal to different data properties. A gradient descent algorithm for parameter optimization is presented which achieves 80.6% recall and 87.7% precision on the FP dataset and 46.5% recall and 79.6% precision on the FPElo dataset. This compares to the results produced with hand-optimization on the test set. Furthermore we investigated the impact of cross-validation and training set selection on recognizer output in order to improve the speech retrieval system.