Product HMMs for audio-visual continuous speech recognition using facial animation parameters. Aleksic, P. & Katsaggelos, A. In 2003 International Conference on Multimedia and Expo. ICME '03. Proceedings (Cat. No.03TH8698), volume 2, pages II–481, 2003. IEEE, IEEE. ![Product HMMs for audio-visual continuous speech recognition using facial animation parameters [link]](https://bibbase.org/img/filetypes/link.svg "link")
Paper doi abstract bibtex The use of visual information in addition to acoustic can improve automatic speech recognition. In this paper we compare different approaches for audio-visual information integration and show how they affect automatic speech recognition performance. We utilize facial animation parameters (FAPs), supported by the MPEG-4 standard for the visual representation as visual features. We use both single-stream and multi-stream hidden Markov models (HMM) to integrate audio and visual information. We performed both state and phone synchronous multi-stream integration. Product HMM topology is used to model the phone-synchronous integration. ASR experiments were performed under noisy audio conditions using a relatively large vocabulary (approximately 1000 words) audio-visual database. The proposed phone-synchronous system, which performed the best, reduces the word error rate (WER) by approximately 20% relatively to audio-only ASR (A-ASR) WERs, at various SNRs with additive white Gaussian noise.
@inproceedings{aleksic2003product,
abstract = {The use of visual information in addition to acoustic can improve automatic speech recognition. In this paper we compare different approaches for audio-visual information integration and show how they affect automatic speech recognition performance. We utilize facial animation parameters (FAPs), supported by the MPEG-4 standard for the visual representation as visual features. We use both single-stream and multi-stream hidden Markov models (HMM) to integrate audio and visual information. We performed both state and phone synchronous multi-stream integration. Product HMM topology is used to model the phone-synchronous integration. ASR experiments were performed under noisy audio conditions using a relatively large vocabulary (approximately 1000 words) audio-visual database. The proposed phone-synchronous system, which performed the best, reduces the word error rate (WER) by approximately 20% relatively to audio-only ASR (A-ASR) WERs, at various SNRs with additive white Gaussian noise.},
author = {Aleksic, P.S. and Katsaggelos, A.K.},
booktitle = {2003 International Conference on Multimedia and Expo. ICME '03. Proceedings (Cat. No.03TH8698)},
doi = {10.1109/ICME.2003.1221658},
isbn = {0-7803-7965-9},
issn = {1945788X},
organization = {IEEE},
pages = {II--481},
publisher = {IEEE},
title = {{Product HMMs for audio-visual continuous speech recognition using facial animation parameters}},
url = {http://ieeexplore.ieee.org/document/1221658/},
volume = {2},
year = {2003}
}
Downloads: 0
{"_id":"m9Kh9DmngFAGaEgWa","bibbaseid":"aleksic-katsaggelos-producthmmsforaudiovisualcontinuousspeechrecognitionusingfacialanimationparameters-2003","author_short":["Aleksic, P.","Katsaggelos, A."],"bibdata":{"bibtype":"inproceedings","type":"inproceedings","abstract":"The use of visual information in addition to acoustic can improve automatic speech recognition. In this paper we compare different approaches for audio-visual information integration and show how they affect automatic speech recognition performance. We utilize facial animation parameters (FAPs), supported by the MPEG-4 standard for the visual representation as visual features. We use both single-stream and multi-stream hidden Markov models (HMM) to integrate audio and visual information. We performed both state and phone synchronous multi-stream integration. Product HMM topology is used to model the phone-synchronous integration. ASR experiments were performed under noisy audio conditions using a relatively large vocabulary (approximately 1000 words) audio-visual database. The proposed phone-synchronous system, which performed the best, reduces the word error rate (WER) by approximately 20% relatively to audio-only ASR (A-ASR) WERs, at various SNRs with additive white Gaussian noise.","author":[{"propositions":[],"lastnames":["Aleksic"],"firstnames":["P.S."],"suffixes":[]},{"propositions":[],"lastnames":["Katsaggelos"],"firstnames":["A.K."],"suffixes":[]}],"booktitle":"2003 International Conference on Multimedia and Expo. ICME '03. Proceedings (Cat. No.03TH8698)","doi":"10.1109/ICME.2003.1221658","isbn":"0-7803-7965-9","issn":"1945788X","organization":"IEEE","pages":"II–481","publisher":"IEEE","title":"Product HMMs for audio-visual continuous speech recognition using facial animation parameters","url":"http://ieeexplore.ieee.org/document/1221658/","volume":"2","year":"2003","bibtex":"@inproceedings{aleksic2003product,\nabstract = {The use of visual information in addition to acoustic can improve automatic speech recognition. In this paper we compare different approaches for audio-visual information integration and show how they affect automatic speech recognition performance. We utilize facial animation parameters (FAPs), supported by the MPEG-4 standard for the visual representation as visual features. We use both single-stream and multi-stream hidden Markov models (HMM) to integrate audio and visual information. We performed both state and phone synchronous multi-stream integration. Product HMM topology is used to model the phone-synchronous integration. ASR experiments were performed under noisy audio conditions using a relatively large vocabulary (approximately 1000 words) audio-visual database. The proposed phone-synchronous system, which performed the best, reduces the word error rate (WER) by approximately 20% relatively to audio-only ASR (A-ASR) WERs, at various SNRs with additive white Gaussian noise.},\nauthor = {Aleksic, P.S. and Katsaggelos, A.K.},\nbooktitle = {2003 International Conference on Multimedia and Expo. ICME '03. Proceedings (Cat. No.03TH8698)},\ndoi = {10.1109/ICME.2003.1221658},\nisbn = {0-7803-7965-9},\nissn = {1945788X},\norganization = {IEEE},\npages = {II--481},\npublisher = {IEEE},\ntitle = {{Product HMMs for audio-visual continuous speech recognition using facial animation parameters}},\nurl = {http://ieeexplore.ieee.org/document/1221658/},\nvolume = {2},\nyear = {2003}\n}\n","author_short":["Aleksic, P.","Katsaggelos, A."],"key":"aleksic2003product","id":"aleksic2003product","bibbaseid":"aleksic-katsaggelos-producthmmsforaudiovisualcontinuousspeechrecognitionusingfacialanimationparameters-2003","role":"author","urls":{"Paper":"http://ieeexplore.ieee.org/document/1221658/"},"metadata":{"authorlinks":{}}},"bibtype":"inproceedings","biburl":"https://sites.northwestern.edu/ivpl/files/2023/06/IVPL_Updated_publications-1.bib","dataSources":["KTWAakbPXLGfYseXn","ePKPjG8C6yvpk4mEK","37Qfzv6wRptYkoSCL","zFPgsTDAW8aDnb5iN","E6Bth2QB5BYjBMZE7","nbnEjsN7MJhurAK9x","PNQZj6FjzoxxJk4Yi","7FpDWDGJ4KgpDiGfB","bod9ms4MQJHuJgPpp","QR9t5P2cLdJuzhfzK","D8k2SxfC5dKNRFgro","7Dwzbxq93HWrJEhT6","qhF8zxmGcJfvtdeAg","fvDEHD49E2ZRwE3fb","H7crv8NWhZup4d4by","DHqokWsryttGh7pJE","vRJd4wNg9HpoZSMHD","sYxQ6pxFgA59JRhxi","w2WahSbYrbcCKBDsC","XasdXLL99y5rygCmq","3gkSihZQRfAD2KBo3","t5XMbyZbtPBo4wBGS","bEpHM2CtrwW2qE8FP","teJzFLHexaz5AQW5z"],"keywords":[],"search_terms":["product","hmms","audio","visual","continuous","speech","recognition","using","facial","animation","parameters","aleksic","katsaggelos"],"title":"Product HMMs for audio-visual continuous speech recognition using facial animation parameters","year":2003}