End-to-End Speech Endpoint Detection Utilizing Acoustic and Language Modeling Knowledge for Online Low-Latency Speech Recognition.
Inyoung HwangJoon-Hyuk ChangPublished in: IEEE Access (2020)
Keyphrases
- neural network
- speech recognition
- end to end
- language model
- language modeling
- high bandwidth
- speech recognition systems
- low latency
- speaker independent
- endpoint detection
- speech recognizers
- pattern recognition
- speech signal
- word error rate
- n gram
- acoustic models
- automatic speech recognition
- noisy speech
- speech synthesis
- retrieval model
- probabilistic model
- speech recognizer
- document retrieval
- query expansion
- information retrieval
- real time
- test collection
- speaker identification
- cross lingual
- hidden markov models
- noisy environments
- multipath
- highly efficient
- ad hoc networks
- feature extraction
- multimedia content
- high throughput
- mixture model
- high speed