Reranked aligners for interactive transcript correction

Abstract : Clarification dialogs can help address ASR errors in speech-to-speech translation systems and other interactive applications. We propose to use variants of Levenshtein alignment for merging an errorful utterance with a targeted rephrase of an error segment. ASR errors that might harm the alignment are addressed through phonetic matching, and a word embedding distance is used to account for the use of synonyms outside targeted segments. These features lead to a relative improvement of 30% of word error rate on ASR output compared to not performing the clarification. Twice as many utterance are completely corrected compared to using basic word alignment. Furthermore, we generate a set of potential merges and train a neural network on crowd-sourced rephrases in order to select the best merger, leading to 24% more instances completely corrected. The system is deployed in the framework of the BOLT project.
Document type :
Conference papers
Complete list of metadatas

https://hal-amu.archives-ouvertes.fr/hal-01194237
Contributor : Benoit Favre <>
Submitted on : Saturday, September 5, 2015 - 11:12:38 AM
Last modification on : Saturday, March 23, 2019 - 1:22:26 AM

Identifiers

  • HAL Id : hal-01194237, version 1

Citation

Benoit Favre, Mickael Rouvier, Frédéric Béchet. Reranked aligners for interactive transcript correction. ICASSP2014 - Speech and Language Processing (ICASSP2014 - SLTC), 2014, Florence, Italy. ⟨hal-01194237⟩

Share

Metrics

Record views

205