# Exploring Deep Hybrid Tensor-to-Vector Network Architectures for Regression Based Speech Enhancement

arXiv: https://arxiv.org/abs/2007.13024 · HTML: https://mlsanity.com/paper/2007.13024/

## Key facts

- Cited by 6 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 25 Jul 2020.

## Built on by (most cited first, 6 of 6)

- [A Study on Speech Enhancement Based on Diffusion Probabilistic Model](https://mlsanity.com/paper/2107.11876/) (2021)
- [QTN-VQC: An End-to-End Learning framework for Quantum Neural Networks](https://mlsanity.com/paper/2110.03861/) (2021)
- [Exploiting Low-Rank Tensor-Train Deep Neural Networks Based on Riemannian Gradient Descent With Illustrations of Speech Processing](https://mlsanity.com/paper/2203.06031/) (2022)
- [Learning Filterbanks for End-to-End Acoustic Beamforming](https://mlsanity.com/paper/2111.04614/) (2021)
- [Exploiting Hybrid Models of Tensor-Train Networks for Spoken Command Recognition](https://mlsanity.com/paper/2201.10609/) (2022)
- [SEOFP-NET: Compression and Acceleration of Deep Neural Networks for Speech Enhancement Using Sign-Exponent-Only Floating-Points](https://mlsanity.com/paper/2111.04436/) (2021)

## Abstract (arXiv)

This paper investigates different trade-offs between the number of model parameters and enhanced speech qualities by employing several deep tensor-to-vector regression models for speech enhancement. We find that a hybrid architecture, namely CNN-TT, is capable of maintaining a good quality performance with a reduced model parameter size. CNN-TT is composed of several convolutional layers at the bottom for feature extraction to improve speech quality and a tensor-train (TT) output layer on the top to reduce model parameters. We first derive a new upper bound on the generalization power of the convolutional neural network (CNN) based vector-to-vector regression models. Then, we provide experimental evidence on the Edinburgh noisy speech corpus to demonstrate that, in single-channel speech enhancement, CNN outperforms DNN at the expense of a small increment of model sizes. Besides, CNN-TT slightly outperforms the CNN counterpart by utilizing only 32% of the CNN model parameters. Besides, further performance improvement can be attained if the number of CNN-TT parameters is increased to 44% of the CNN model size. Finally, our experiments of multi-channel speech enhancement on a simulated noisy WSJ0 corpus demonstrate that our proposed hybrid CNN-TT architecture achieves better results than both DNN and CNN models in terms of better-enhanced speech qualities and smaller parameter sizes.