Recursions-Are-All-You-Need VS MockingBird

The use of deep unfolding networks in compressive sensing (CS) has seen wide success as they provide both simplicity and interpretability. However, since most deep unfolding networks are iterative, this incurs significant redundancies in the network. In this work, we propose a novel recursion-based framework to enhance the efficiency of deep unfolding models. First, recursions are used to effectively eliminate the redundancies in deep unfolding networks. Secondly, we randomize the number of recursions during training to decrease the overall training time. Finally, to effectively utilize the power of recursions, we introduce a learnable unit to modulate the features of the model based on both the total number of iterations and the current iteration index. To evaluate the proposed framework, we apply it to both ISTA-Net+ and COAST. Extensive testing shows that our proposed framework allows the network to cut down as much as 75% of its learnable parameters while mostly maintaining its performance, and at the same time, it cuts around 21% and 42% from the training time for ISTA-Net+ and COAST respectively. Moreover, when presented with a limited training dataset, the recursive models match or even outperform their respective non-recursive baseline. Codes and pretrained models are available at https://github.com/Rawwad-Alhejaili/Recursions-Are-All-You-Need .

Actually, there are already open source implementations available, for example, the MockingBird project on GitHub. It supports English and Mandarin Chinese. For those with enough computation power and willingness to try, you can even make your own voice dataset and train the model to generate ‘your’ sound, simply following the project docs.

piece of recordings MockingBird

项目地址：https://github.com/babysor/MockingBird

Mocking Bird – Realtime Voice Clone for Chinese\ (45 comments)

I wonder if there is a way to tie in this AI voice changer. https://github.com/babysor/MockingBird

View on GitHub