Pseudo-segment based speech recognition using neural recurrent whole-word recognizers.
Philippe Le CerfKris DemuynckJacques DuchateauDirk Van CompernollePublished in: ICASSP (1) (1994)
Keyphrases
- speech recognizer
- speech recognition
- word recognition
- spoken language
- speech recognizers
- automatic speech recognition
- speech recognition systems
- broadcast news
- network architecture
- spiking neural networks
- speech segments
- speech signal
- neural network
- recognition errors
- recurrent neural networks
- speech synthesis
- hidden markov models
- prosodic features
- english text
- recurrent networks
- noisy environments
- spontaneous speech
- feed forward
- lexical features
- spoken document retrieval
- word error rate
- biologically plausible
- text input
- spoken documents
- speech sounds
- language model
- character recognition
- out of vocabulary
- video sequences
- pattern recognition
- artificial neural networks
- n gram
- biologically inspired
- neural model
- handwriting recognition
- keyword spotting
- dialogue system
- associative memory
- video search
- co occurrence
- speaker identification
- text to speech