K
Kai Zhen
Researcher at Indiana University
Publications - 20
Citations - 136
Kai Zhen is an academic researcher from Indiana University. The author has contributed to research in topics: Codec & Artificial neural network. The author has an hindex of 4, co-authored 20 publications receiving 51 citations. Previous affiliations of Kai Zhen include Xidian University & Electronics and Telecommunications Research Institute.
Papers
More filters
Proceedings ArticleDOI
Cascaded Cross-Module Residual Learning Towards Lightweight End-to-End Speech Coding.
TL;DR: In this paper, a cross-module residual learning (CMRL) pipeline is proposed as a module carrier with each module reconstructing the residual from its preceding modules, which shows better objective performance than AMR-WB and OPUS.
Journal ArticleDOI
Psychoacoustic Calibration of Loss Functions for Efficient End-to-End Neural Audio Coding
TL;DR: This work presents a psychoacoustic calibration scheme to re-define the loss functions of neural audio coding systems so that it can decode signals more perceptually similar to the reference, yet with a much lower model complexity.
Proceedings ArticleDOI
Efficient and Scalable Neural Residual Waveform Coding with Collaborative Quantization
TL;DR: In this paper, a collaborative quantization (CQ) scheme is proposed to jointly learn the codebook of LPC coefficients and the corresponding residuals, which achieves much higher quality than its predecessor at 9 kbps with even lower model complexity.
Posted Content
Cascaded Cross-Module Residual Learning towards Lightweight End-to-End Speech Coding
TL;DR: A cross-module residual learning (CMRL) pipeline as a module carrier with each module reconstructing the residual from its preceding modules in a two-phase training scheme, showing better objective performance than AMR-WB and the state-of-the-art DNN-based speech codec with a similar network architecture.
Proceedings ArticleDOI
Sparsification via Compressed Sensing for Automatic Speech Recognition
TL;DR: In this paper, a compressed sensing based pruning (CSP) approach is proposed to effectively address the questions of when and which weights should be forced to zero, i.e. be pruned.