On time-frequency mask estimation for MVDR beamforming with application in robust speech recognition.
Xiong XiaoShengkui ZhaoDouglas L. JonesEng Siong ChngHaizhou LiPublished in: ICASSP (2017)
Keyphrases
- speech recognition
- noisy environments
- language model
- speech synthesis
- hidden markov models
- speech understanding
- automatic speech recognition
- keyword spotting
- pattern recognition
- speech processing
- estimation error
- speech signal
- frequency domain
- speech recognition technology
- speaker identification
- speaker recognition
- noise reduction
- maximum likelihood
- signal processing
- speech enhancement
- information retrieval
- speech recognition systems
- data mining