{"department":[{"_id":"54"}],"oa":"1","type":"journal_article","citation":{"short":"S. Windmann, R. Haeb-Umbach, 2008 ITG Conference on Voice Communication (SprachKommunikation) (2008) 1–4.","ama":"Windmann S, Haeb-Umbach R. A novel approach to noise estimation in model-based speech feature enhancement. <i>2008 ITG Conference on Voice Communication (SprachKommunikation)</i>. 2008:1-4.","ieee":"S. Windmann and R. Haeb-Umbach, “A novel approach to noise estimation in model-based speech feature enhancement,” <i>2008 ITG Conference on Voice Communication (SprachKommunikation)</i>, pp. 1–4, 2008.","bibtex":"@article{Windmann_Haeb-Umbach_2008, title={A novel approach to noise estimation in model-based speech feature enhancement}, journal={2008 ITG Conference on Voice Communication (SprachKommunikation)}, author={Windmann, Stefan and Haeb-Umbach, Reinhold}, year={2008}, pages={1–4} }","chicago":"Windmann, Stefan, and Reinhold Haeb-Umbach. “A Novel Approach to Noise Estimation in Model-Based Speech Feature Enhancement.” <i>2008 ITG Conference on Voice Communication (SprachKommunikation)</i>, 2008, 1–4.","mla":"Windmann, Stefan, and Reinhold Haeb-Umbach. “A Novel Approach to Noise Estimation in Model-Based Speech Feature Enhancement.” <i>2008 ITG Conference on Voice Communication (SprachKommunikation)</i>, 2008, pp. 1–4.","apa":"Windmann, S., &#38; Haeb-Umbach, R. (2008). A novel approach to noise estimation in model-based speech feature enhancement. <i>2008 ITG Conference on Voice Communication (SprachKommunikation)</i>, 1–4."},"author":[{"first_name":"Stefan","full_name":"Windmann, Stefan","last_name":"Windmann"},{"first_name":"Reinhold","id":"242","full_name":"Haeb-Umbach, Reinhold","last_name":"Haeb-Umbach"}],"year":"2008","user_id":"44006","main_file_link":[{"url":"https://groups.uni-paderborn.de/nt/pubs/2008/WiHa08-2.pdf","open_access":"1"}],"date_created":"2019-07-12T05:31:12Z","_id":"11940","status":"public","publication":"2008 ITG Conference on Voice Communication (SprachKommunikation)","abstract":[{"text":"In this paper, the noise estimation for model-based speech feature enhancement in automatic speech recognition (ASR) is investigated. Beside a stationary noise prior, three linear state space models for the (cepstral) noise process are considered. We have derived novel EM algorithms for the estimation of the noise model parameters: A blockwise EM algorithm is applied on noise-only input data. It is supposed to be used during the offline training mode of the recognizer. Further a sequential online EM algorithm is employed to adapt the observation variance in recognition mode which works as well under the asumption of a stationary noise prior and a linear state model for the noise. Experiments on the AURORA4 database lead to improved recognition results with the new state model compared to the assumption of stationary noise.","lang":"eng"}],"title":"A novel approach to noise estimation in model-based speech feature enhancement","date_updated":"2022-01-06T06:51:12Z","language":[{"iso":"eng"}],"page":"1-4"}