ESPnet-SE++: Speech Enhancement for Robust Speech Recognition, Translation, and Understanding.
Yen-Ju LuXuankai ChangChenda LiWangyou ZhangSamuele CornellZhaoheng NiYoshiki MasuyamaBrian YanRobin ScheiblerZhong-Qiu WangYu TsaoYanmin QianShinji WatanabePublished in: INTERSPEECH (2022)
Keyphrases
- speech recognition
- noisy environments
- speech enhancement
- speech signal
- noisy speech
- automatic speech recognition
- language model
- hidden markov models
- background noise
- speech synthesis
- pattern recognition
- spectral subtraction
- vocal tract
- speaker identification
- noise reduction
- speech recognition systems
- linear prediction
- spectral analysis
- neural network