This paper proposes an instantaneous speaker adaptation method that uses N-best decoding for continuous mixture-density hidden-Markovmodel-based speech-recognition systems. This method is effective even for speakers whose decoding using speaker-independent (SI) models are error-prone and for whom sp
β¦ LIBER β¦
Adaptive systems for unsupervised speaker tracking and speech recognition
β Scribed by Tobias Herbig; Franz Gerl; Wolfgang Minker; Reinhold Haeb-Umbach
- Book ID
- 107661761
- Publisher
- Springer-Verlag
- Year
- 2011
- Tongue
- English
- Weight
- 1005 KB
- Volume
- 2
- Category
- Article
- ISSN
- 1868-6478
No coin nor oath required. For personal study only.
π SIMILAR VOLUMES
N-Best-based unsupervised speaker adapta
β
Tomoko Matsui; Sadaoki Furui
π
Article
π
1998
π
Elsevier Science
π
English
β 251 KB
Speaker adaptation techniques for speech
β
Koichi Shinoda
π
Article
π
2005
π
John Wiley and Sons
π
English
β 353 KB
Speaker adaptation method for fenonic ma
β
Masafumi Nishimura
π
Article
π
1991
π
John Wiley and Sons
π
English
β 845 KB
Predictor codebook for speaker-independe
β
Takeshi Kawabata
π
Article
π
1994
π
John Wiley and Sons
π
English
β 752 KB
## Abstract This paper discusses a method to handle the diversified dynamic features of speech by representing the dynamic features of speech by spectrum predictors and constructing the codebook containing predictors as the elements. The effectiveness of the method for speakerβindependent speech re
Normalization and adaptation of speech d
β
R.W.A. Scarr
π
Article
π
1970
π
Elsevier Science
β 933 KB
Adaptation to a speaker's voice in a spe
Adaptation to a speaker's voice in a speech recognition system based on synthetic phoneme references
β
Mats Blomberg
π
Article
π
1991
π
Elsevier Science
π
English
β 698 KB