Speech enhancement with weighted denoising auto-encoder
By Bingyin Xia, Chang-chun Bao. 2013
63 citations recorded by the source corpus.
Subfields: Artificial Intelligence, Computer Vision and Pattern Recognition, Signal Processing
How researchers see this paper
The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.
In Artificial Intelligence
- Shinji Watanabe
- Björn Schuller
- Zixing Zhang
- Felix Weninger
- Jürgen T. Geiger
- Chin-Hui Lee
- Amr El-Desoky Mousa
- Jouni Pohjalainen
In Computer Vision and Pattern Recognition
- Felix Weninger
- Jean-Marc Valin
- Björn Schuller
- Jun Du
- Ritwik Giri
- Umut Isik
- Zixing Zhang
- Shinji Watanabe
In Signal Processing
- Shinji Watanabe
- Jean-Marc Valin
- John R. Hershey
- Deliang Wang
- Chin-Hui Lee
- Jun Du
- Björn Schuller
- Zixing Zhang
Citations in the corpus
- Enhancing into the Codec: Noise Robust Speech Coding with Vector-Quantized Autoencoders (2021, methodology): “Motivated by the performance of modern neural speech enhancers in removing unwanted noise and reverberation from audio signals [13, 14, 15, 16, 17], we combine the learning based compression and learning based enhancement paradigms.”
- Deep Learning for Environmentally Robust Speech Recognition (2017, methodology): “nd the log-Mel-spectral domains [36], respectively. Motivated by the fact that the same distortion in different frequency bands has different effects on speech quality, a weighted SAE was proposed in [64] and showed positive performancefor denoising.In detail, a weighted reconstruction loss function”
- PoCoNet: Better Speech Enhancement with Frequency-Positional Embeddings, Semi-Supervised Conversational Data, and Biased Loss (2020, background): “Neural network based approaches have greatly improved the output quality of speech enhancement systems [1, 2, 3, 4, 5].”
- A Regression Approach to Speech Enhancement Based on Deep Neural Networks (2015, methodology): “In [22, 23], stacked denoising autoencoders (SDAs), as one type of the deep models, were adopted to model the relationship between clean and noisy features, and they only explored its performance on a matching test set.”
- Deep recurrent de-noising auto-encoder and blind de-reverberation for reverberated speech recognition (2014, background): “DAE-based feature enhancement has been considered by [3, 18], but these studies only consider additive, not convolutional, noise.”