Dual-Path RNN for Long Recording Speech Separation.
Chenda LiYi LuoCong HanJinyu LiTakuya YoshiokaTianyan ZhouMarc DelcroixKeisuke KinoshitaChristoph BöddekerYanmin QianShinji WatanabeZhuo ChenPublished in: SLT (2021)
Keyphrases
- recurrent neural networks
- nearest neighbor
- speech recognition
- shortest path
- speech signal
- text to speech
- neural network
- broadcast news
- spoken language
- optimal path
- automatic speech recognition
- user interface
- knn
- hidden markov models
- emotion recognition
- lower bound
- face recognition
- data sets
- recognition engine
- text to speech synthesis