Theme Identification in Telephone Service Conversations using Quaternions of Speech Features

Abstract : The paper introduces new features for describing possible focus variation in a human/human conversation. The application considered is a real-life telephone customer care service. The purpose is to hypothesize the dominant theme of conversations between a casual customer calling. Conversations are processed by an automatic speech recognition system that provides hypotheses used for extracting word frequency. Features are extracted in different, broadly defined and partially overlapped, time segments. Combinations of each feature in different segments are represented in a quaternion algebra framework. The advantage of the proposed approach is made evident by the statistically significant improvements in theme classification accuracy .
Document type :
Conference papers
Complete list of metadatas

https://hal.archives-ouvertes.fr/hal-01339930
Contributor : Bibliothèque Universitaire Déposants Hal-Avignon <>
Submitted on : Thursday, June 30, 2016 - 10:42:39 AM
Last modification on : Saturday, March 23, 2019 - 1:22:32 AM

Identifiers

  • HAL Id : hal-01339930, version 1

Collections

Citation

Mohamed Morchid, Georges Linarès, Marc El Bèze, Renato de Mori. Theme Identification in Telephone Service Conversations using Quaternions of Speech Features. Interspeech 2013, Aug 2013, Lyon, France. ⟨hal-01339930⟩

Share

Metrics

Record views

66