Directional ASR: A New Paradigm for E2E Multi-Speaker Speech Recognition with Source Localization.
Aswin Shanmugam SubramanianChao WengShinji WatanabeMeng YuYong XuShi-Xiong ZhangDong YuPublished in: CoRR (2020)
Keyphrases
- speech recognition
- source localization
- automatic speech recognition
- speech signal
- hidden markov models
- wireless sensor networks
- language model
- noisy environments
- word error rate
- speech processing
- speaker identification
- sound source
- pattern recognition
- speaker dependent
- speech synthesis
- speaker recognition
- speech recognizer
- speaker independent
- handwriting recognition
- speech retrieval
- speech recognition technology
- speech recognition systems
- broadcast news
- word recognition
- speaker diarization
- speaker adaptation
- neural network
- speech recognizers
- acoustic models
- conversational speech
- vocal tract
- cepstral coefficients
- acoustic features
- dynamic environments
- denoising
- mobile devices