Prediction of transcription indexability

Abstract : Prediction of transcription indexability This paper presents a semantic confidence measure that aims to predict the relevance of automatic transcripts for a task of Spoken Document Retrieval (SDR). The proposed predicting method relies on the combination of Automatic Speech Recognition confidence measure and a Semantic Compacity Index, that estimates the relevance of the words considering the semantic context in which they occurred. Experiments are conducted on the French Broadcast news corpus ESTER 2, by simulating a classical SDR usage scenario : users submit text-queries to a search engine that is expected to return the most relevant documents regarding the query. Results demonstrate the interest of using semantic level information to predict the transcription indexability.
Document type :
Conference papers
Liste complète des métadonnées

Cited literature [12 references]  Display  Hide  Download
Contributor : Benjamin Lecouteux <>
Submitted on : Thursday, November 23, 2017 - 10:14:11 AM
Last modification on : Tuesday, February 12, 2019 - 1:31:30 AM


Publisher files allowed on an open archive


  • HAL Id : hal-00954215, version 1



Gregory Senay, Benjamin Lecouteux, Georges Linares. Prediction of transcription indexability. Actes de la conférence conjointe JEP-TALN-RECITAL 2012, volume 1: JEP, 2012, Grenoble, France. pp.x-x. ⟨hal-00954215⟩



Record views


Files downloads