Training an automatic speech recognition system using compressed word frequencies
作者： Mitchel Weintraub , Brian Strope
DOI:
关键词:
摘要: Respective word frequencies may be determined from a corpus of utterance-to-text-string mappings that contain associations between audio utterances and respective text string transcription each utterance. compressed obtained based on the such distribution has lower variance than frequencies. Sample selected An automatic speech recognition (ASR) system trained with sample mappings.
google.co.uk 参考文章(147)
Joshua T. Goodman, A Bit of Progress in Language Modeling Extended Version Microsoft Research. pp. 72- ,(2001)
PC Woodland, Speaker adaptation for continuous density HMMs: a review ISCA: International Speech Communication Association. ,(2001)
Mehryar Mohri, Weighted Finite-State Transducer Algorithms. An Overview Formal Languages and Applications. pp. 551- 563 ,(2004) , 10.1007/978-3-540-39886-8_29
Tu Van Le, Dat Tran, Michael Wagner, Fuzzy Gaussian mixture models for speaker recognition. conference of the international speech communication association. ,(1998)
Xiaodong He, Jonathan Hamaker, Xin Lei, Patrick Nguyen, Speech recognition using adaptation and prior knowledge ,(2005)
Etienne Marcheret, Yuqing Gao, Hakan Erdogan, Yongxin Li, Incremental on-line feature space MLLR adaptation for telephony speech recognition. conference of the international speech communication association. ,(2002)
Xiaodong He, Xin Lei, Jon Hamaker, Robust feature space adaptation for telephony speech recognition. conference of the international speech communication association. ,(2006)
Toru Imai, Richard M. Schwartz, Topic indexing method ,(1998)
Michael Riley, Fernando Pereira, Andrej Ljolje, Efficient general lattice generation and rescoring. conference of the international speech communication association. ,(1999)
Mark Epstein, Hierarchical language models ,(2001)