Speech recognition in a realistic noisy environment using multiple microphones is the focal point of the third CHiME challenge. Over the baseline ASR system provided for this challenge, we apply state of the art algorithms for boosting acoustic model learning and hypothesis rescoring to improve the final output. To this aim, we first use the automatic transcription of each channel to re-train the acoustic model for that channel and then we apply linear language model rescoring to find a better solution in the n-best list. LM rescoring is performed using an efficient set of N-gram and Recurrent Neural Network LM (RNNLM) trained on a wisely-selected text set. In the experiments, we show that the proposed approach improves not only the individual channel transcription, but also the enhanced channels produced by MVDR and delay-and-sum beamforming.
Boosted acoustic model learning and hypotheses rescoring on the CHiME-3 task
Jalalvand, Shahab;Falavigna, Giuseppe Daniele;Matassoni, Marco;Svaizer, Piergiorgio;Omologo, Maurizio
2015-01-01
Abstract
Speech recognition in a realistic noisy environment using multiple microphones is the focal point of the third CHiME challenge. Over the baseline ASR system provided for this challenge, we apply state of the art algorithms for boosting acoustic model learning and hypothesis rescoring to improve the final output. To this aim, we first use the automatic transcription of each channel to re-train the acoustic model for that channel and then we apply linear language model rescoring to find a better solution in the n-best list. LM rescoring is performed using an efficient set of N-gram and Recurrent Neural Network LM (RNNLM) trained on a wisely-selected text set. In the experiments, we show that the proposed approach improves not only the individual channel transcription, but also the enhanced channels produced by MVDR and delay-and-sum beamforming.File | Dimensione | Formato | |
---|---|---|---|
CHiME3_ASRU2015.pdf
solo utenti autorizzati
Descrizione: articolo
Tipologia:
Documento in Post-print
Licenza:
NON PUBBLICO - Accesso privato/ristretto
Dimensione
267.34 kB
Formato
Adobe PDF
|
267.34 kB | Adobe PDF | Visualizza/Apri Richiedi una copia |
I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.