CINECA IRIS Institutional Research Information System
IRIS è l’archivio della produzione scientifica della Fondazione Bruno Kessler.
EnglishThis work presents acoustic model adaptation based on a piece-wise energy decay curve. The dual slope representation of the piece-wise curve accurately captures the early and late reflection decay which helps in precisely modeling the smearing effect caused due to reverberation. Adaptation using piece-wise decay curve leads to robust acoustic models consequently improving the recognition performance. The approach is tested on connected digits recognition task in different rooms with various reverberation times. The performance is compared with the exponential decay approach and incremental MLLR, where the proposed approach was more effective. Moreover, the combination of piece-wise adaptation with incremental MLLR is also studied and the combination is instrumental in improving the performance with respect to incremental MLLR.
Scheda prodotto non validato
Attenzione! I dati visualizzati non sono stati sottoposti a validazione da parte di FBK.
| Titolo: | Acoustic Model Adaptation Using Piece-wise Energy Decay Curve for Reverberant Environments | |
| Autori: | ||
| Data di pubblicazione: | 2012 | |
| Abstract: | This work presents acoustic model adaptation based on a piece-wise energy decay curve. The dual slope representation of the piece-wise curve accurately captures the early and late reflection decay which helps in precisely modeling the smearing effect caused due to reverberation. Adaptation using piece-wise decay curve leads to robust acoustic models consequently improving the recognition performance. The approach is tested on connected digits recognition task in different rooms with various reverberation times. The performance is compared with the exponential decay approach and incremental MLLR, where the proposed approach was more effective. Moreover, the combination of piece-wise adaptation with incremental MLLR is also studied and the combination is instrumental in improving the performance with respect to incremental MLLR. | |
| Handle: | http://hdl.handle.net/11582/104209 | |
| Appare nelle tipologie: | 4.1 Contributo in Atti di convegno |
File in questo prodotto:
http://hdl.handle.net/11582/104209
Copyright © 2021