Service interruption on Monday 11 July from 12:30 to 13:00: all the sites of the CCSD (HAL, Epiciences, SciencesConf, AureHAL) will be inaccessible (network hardware connection).
Skip to Main content Skip to Navigation
Conference papers

Réordonnancement automatique d'hypothèses pour l'assistancè a la transcription de la parole

Abstract : Large vocabulary automatic speech recognition (ASR) technologies perform well in known, controlled contexts. However, some mistakes still have to be corrected. Human intervention is necessary to check and correct the results of such systems in order to make the output of ASR understandable. We propose a method for computer-assisted transcription of speech, based on automatic reordering confusion networks. It allows to significantly reduce the number of actions needed to correct the ASR outputs. WER computed before and after every network reordering shows an absolute gain of about 3.4%.
Document type :
Conference papers
Complete list of metadata

Cited literature [12 references]  Display  Hide  Download

https://hal.archives-ouvertes.fr/hal-01433897
Contributor : sylvain meignier Connect in order to contact the contributor
Submitted on : Monday, April 3, 2017 - 10:11:53 PM
Last modification on : Thursday, April 6, 2017 - 10:05:50 AM
Long-term archiving on: : Tuesday, July 4, 2017 - 2:54:43 PM

File

Laurent10j1.pdf
Publisher files allowed on an open archive

Identifiers

  • HAL Id : hal-01433897, version 1

Collections

Citation

Antoine Laurent, Sylvain Meignier, Paul Deléglise. Réordonnancement automatique d'hypothèses pour l'assistancè a la transcription de la parole. XXVIIIe Journées d'étude sur la parole (JEP 2010), 2010, Mons, Belgique. ⟨hal-01433897⟩

Share

Metrics

Record views

102

Files downloads

16