The search result changed since you submitted your search request. Documents might be displayed in a different sort order.
  • search hit 126 of 660
Back to Result List

An uncertainty decoding approach to noise- and reverberation-robust speech recognition

  • The generic REMOS (REverberation MOdeling for robust Speech recognition) concept is extended in this contribution to cope with additional noise components. REMOS originally embeds an explicit reverberation model into a hiddenMarkov model (HMM) leading to a relaxed conditional independence assumption for the observed feature vectors. During recognition, a nonlinear optimization problem is to be solved in order to adapt the HMMs' output probability density functions to the current reverberation conditions. The extension for additional noise components necessitates a modified numerical solver for the nonlinear optimization problem. We propose an approximation scheme based on continuous piecewise linear regression. Connected-digit recognition experiments demonstrate the potential of REMOS in reverberant and noisy environments. They furthermore reveal that the benefit of an explicit reverberation model, overcoming the conditional independence assumption, increases with increasing signal-to-noise-ratios.

Export metadata

Additional Services

Share in Twitter Search Google Scholar Statistics
Metadaten
Author:Roland Maas, Akshaya Thippur, Armin Sehr, Walter Kellermann
DOI:https://doi.org/10.1109/ICASSP.2013.6639098
Parent Title (English):IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), 26-31 May 2013, Vancouver, BC, Canada : Proceedings
Publisher:IEEE
Document Type:conference proceeding (article)
Language:English
Year of first Publication:2013
Release Date:2022/11/25
Tag:Adaptation models; Hidden Markov models; Noise; Optimization; Reverberation; Speech; Vectors
First Page:7388
Last Page:7392
Institutes:Fakultät Elektro- und Informationstechnik
Fakultät Elektro- und Informationstechnik / Labor Elektroakustik
Begutachtungsstatus:peer-reviewed
Publication:Externe Publikationen
research focus:Digitalisierung
Licence (German):Keine Lizenz - Es gilt das deutsche Urheberrecht: § 53 UrhG
Einverstanden ✔
Diese Webseite verwendet technisch erforderliche Session-Cookies. Durch die weitere Nutzung der Webseite stimmen Sie diesem zu. Unsere Datenschutzerklärung finden Sie hier.