Topic-space based setup of a neural network for theme identification of highly imperfect transcriptions

Abstract : This paper presents a method for speech analytics that integrates topic-space based representation into a feed-forward artificial neural network (FFANN), working as a document classifier. The proposed method consists in configuring the FFANN's topology and in initializing the weights according to a previously estimated topic-space. Setup based on thematic priors is expected to improve the efficiency of the FFANN's weight optimization process, while speeding-up the training process and improving the classification accuracy. This method is evaluated on a spoken dialogue categorization task which is composed of customer-agent dialogues from the call-centre of Paris Public Transportation Company. Results show the interest of the proposed setup method, with a gain of more than 4 points in terms of classification accuracy, compared to the baseline. Moreover, experiments highlight that performance is weakly dependent to FFANN's topology with the LDA-based configuration, in comparison to classical empirical setup.
Document type :
Conference papers
Complete list of metadatas

https://hal.archives-ouvertes.fr/hal-01339956
Contributor : Bibliothèque Universitaire Déposants Hal-Avignon <>
Submitted on : Thursday, June 30, 2016 - 10:52:29 AM
Last modification on : Saturday, March 23, 2019 - 1:22:34 AM

Identifiers

Collections

Citation

Mohamed Morchid, Richard Dufour, Georges Linarès. Topic-space based setup of a neural network for theme identification of highly imperfect transcriptions. 2015 IEEE Workshop on Automatic Speech Recognition and Understanding (ASRU) , Dec 2015, Scottsdale, United States. ⟨10.1109/ASRU.2015.7404815⟩. ⟨hal-01339956⟩

Share

Metrics

Record views

72