A 1000-word vocabulary, speaker-independent, continuous live-mode speech recognizer implemented in a single FPGA.
Edward C. LinKai YuRob A. RutenbarTsuhan ChenPublished in: FPGA (2007)
Keyphrases
- speech recognizer
- speaker independent
- speech recognition
- spoken language
- speech recognition systems
- out of vocabulary
- broadcast news
- speaker dependent
- acoustic models
- language model
- speaker adaptation
- hidden markov models
- speaker identification
- keywords
- automatic speech recognition
- cross language information retrieval
- n gram
- video data
- natural language
- pattern recognition
- neural network