F-T-LSTM Based Complex Network for Joint Acoustic Echo Cancellation and Speech Enhancement.
Shimin ZhangYuxiang KongShubo LvYanxin HuLei XiePublished in: Interspeech (2021)
Keyphrases
- complex networks
- speech enhancement
- noisy speech
- noisy environments
- noise reduction
- signal to noise ratio
- speech signal
- single channel
- scale free
- speech recognition
- social networks
- social network analysis
- network structure
- graph theory
- hidden markov models
- community structure
- maximal cliques
- background noise
- acoustic features
- information theoretic concepts
- clustering coefficient
- wiener filter
- linear prediction
- smoothing algorithm
- edge preserving
- sound source
- small world
- additive noise
- vocal tract
- image restoration
- edge detection