Senone scoring for multiple input streams

作者: Ojas A. Bapat

DOI:

关键词:

摘要: Embodiments of the present invention include an apparatus, method, and system for calculating senone scores multiple concurrent input speech streams. The method can following: receiving one or more feature vectors from streams; accessing acoustic model at a time; separate corresponding to each incoming vector. calculation uses single read access calculates set vectors, before proceeding next in model.

参考文章(11)
Vyacheslav A. Barannikov, Alexandr A. Kibkalo, Method and apparatus for dynamic beam control in viterbi search ,(2001)
Li Jiang, Xuedong Huang, Speech recognition method and apparatus utilizing multiple feature streams ASAJ. ,vol. 114, pp. 1208- 1208 ,(1999)
Yoanna Baumgartner, Speech label accelerators and techniques for using same Journal of the Acoustical Society of America. ,vol. 121, pp. 29- ,(2001) , 10.1121/1.2434333
Raymond Hoare, Kshitij Gupta, Jeffrey W. Schuster, Speech silicon AM: an FPGA-based acoustic modeling pipeline for hidden Markov model based speech recognition international parallel and distributed processing symposium. pp. 210- 210 ,(2006) , 10.5555/1898699.1898700
Anthony Chun, Jenny X. Chang, Zhen Fang, Ravishankar Iyer, Michael Deisher, ISIS: An accelerator for Sphinx speech recognition symposium on application specific processors. pp. 58- 61 ,(2011) , 10.1109/SASP.2011.5941078
Ullas Pazhayaveetil, Dhruba Chandra, Paul Franzon, Flexible Low Power Probability Density Estimation Unit For Speech Recognition international symposium on circuits and systems. pp. 1117- 1120 ,(2007) , 10.1109/ISCAS.2007.378206
Gerasimos Potamianos, Etienne Marcheret, Vaibhava Goel, Stephen Mingyu Chu, Method for likelihood computation in multi-stream HMM based speech recognition ,(2004)