An End-to-End Approach to Joint Social Signal Detection and Automatic Speech Recognition.
Hirofumi InagumaMasato MimuraKoji InoueKazuyoshi YoshiiTatsuya KawaharaPublished in: ICASSP (2018)
Keyphrases
- end to end
- automatic speech recognition
- signal detection
- speech recognition
- speech signal
- broadcast news
- word error rate
- hidden markov models
- conversational speech
- weak signal
- speech retrieval
- admission control
- congestion control
- spontaneous speech
- low signal to noise ratio
- spoken words
- word recognition
- internet protocol
- recognition errors
- language model
- computer vision
- application layer
- wireless ad hoc networks
- error rate
- speech corpus
- compound words
- scalable video
- pattern recognition
- speech sounds
- machine learning