ABSTRACT: Automatic transcription of spoken documents is affected by automatic transcription errors that are especially frequent when speech is acquired in severe noisy conditions. Automatic Speech Recognition errors induce errors in the linguistic features used for a variety of Natural Language Processing tasks. Recently, denoising autoencoders and stacked autoencoders have been proposed with interesting results for acoustic feature denoising tasks. This talk will deal with the recovery of corrupted linguistic features in spoken documents. Solutions based on denoising and stacked autoencoders will be considered and evaluated in a spoken conversation analysis task.

BIO: https://www.researchgate.net/profile/Renato_De_Mori