116 citations · 146 across the 16 of their papers we have counts for
7 papers · 1 filter
Enhance audio generation controllability through representation similarity regularization
Yangyang Shi, Gael Le Lan, Varun Nagaraja +6
This paper presents an innovative approach to enhance control over audio generation by emphasizing the alignment between audio and text representations during model training. In th…
Ripple sparse self-attention for monaural speech enhancement
Qiquan Zhang, Hongxu Zhu, Qi Song +3
The use of Transformer represents a recent success in speech enhancement. However, as its core component, self-attention suffers from quadratic complexity, which is computationally…
ESPnet-ST-v2: Multipurpose Spoken Language Translation Toolkit
Brian Yan, Jiatong Shi, Yun Tang +13
ESPnet-ST-v2 is a revamp of the open-source ESPnet-ST toolkit necessitated by the broadening interests of the spoken language translation community. ESPnet-ST-v2 supports 1) offlin…
Time-Frequency Attention for Monaural Speech Enhancement
Qiquan Zhang, Qi Song, Zhaoheng Ni +2
Most studies on speech enhancement generally don't consider the energy distribution of speech in time-frequency (T-F) representation, which is important for accurate prediction of…
Combining Spatial Clustering with LSTM Speech Models for Multichannel Speech Enhancement
Felix Grezes, Zhaoheng Ni, Viet Anh Trinh +1
Recurrent neural networks using the LSTM architecture can achieve significant single-channel noise reduction. It is not obvious, however, how to apply them to multi-channel inputs…
Improved MVDR Beamforming Using LSTM Speech Models to Clean Spatial Clustering Masks
Zhaoheng Ni, Felix Grezes, Viet Anh Trinh +1
Spatial clustering techniques can achieve significant multi-channel noise reduction across relatively arbitrary microphone configurations, but have difficulty incorporating a detai…
Enhancement of Spatial Clustering-Based Time-Frequency Masks using LSTM Neural Networks
Felix Grezes, Zhaoheng Ni, Viet Anh Trinh +1
Recent works have shown that Deep Recurrent Neural Networks using the LSTM architecture can achieve strong single-channel speech enhancement by estimating time-frequency masks. How…