We investigate the problem of predicting the quality of automatic speech recognition (ASR) output under the following rigid constraints: i) reference transcriptions are not available, ii) confidence information about the system that produced the transcriptions is not accessible, and iii) training and test data come from multiple domains. To cope with these constraints (typical of the constantly increasing amount of automatic transcriptions that can be found on the Web), we propose a domain-adaptive approach based on multitask learning. Different algorithms and strategies are evaluated with English data coming from four domains, showing that the proposed approach can cope with the limitations of previously proposed single task learning methods.

Multitask Learning for Adaptive Quality Estimation of Automatically Transcribed Utterances

Negri, Matteo;Turchi, Marco;Falavigna, Giuseppe Daniele
2015-01-01

Abstract

We investigate the problem of predicting the quality of automatic speech recognition (ASR) output under the following rigid constraints: i) reference transcriptions are not available, ii) confidence information about the system that produced the transcriptions is not accessible, and iii) training and test data come from multiple domains. To cope with these constraints (typical of the constantly increasing amount of automatic transcriptions that can be found on the Web), we propose a domain-adaptive approach based on multitask learning. Different algorithms and strategies are evaluated with English data coming from four domains, showing that the proposed approach can cope with the limitations of previously proposed single task learning methods.
2015
978-1-941643-49-5
File in questo prodotto:
File Dimensione Formato  
naacl2015-MTL.pdf

accesso aperto

Tipologia: Documento in Post-print
Licenza: Creative commons
Dimensione 625.49 kB
Formato Adobe PDF
625.49 kB Adobe PDF Visualizza/Apri

I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.

Utilizza questo identificativo per citare o creare un link a questo documento: https://hdl.handle.net/11582/300441
Citazioni
  • ???jsp.display-item.citation.pmc??? ND
social impact