Picture for Kai Zhen

Kai Zhen

Indiana University Bloomington, Alexa Machine Learning, Amazon, USA

AdaZeta: Adaptive Zeroth-Order Tensor-Train Adaption for Memory-Efficient Large Language Models Fine-Tuning

Add code
Jun 26, 2024
Viaarxiv icon

Sub-8-bit quantization for on-device speech recognition: a regularization-free approach

Add code
Oct 17, 2022
Figure 1 for Sub-8-bit quantization for on-device speech recognition: a regularization-free approach
Figure 2 for Sub-8-bit quantization for on-device speech recognition: a regularization-free approach
Figure 3 for Sub-8-bit quantization for on-device speech recognition: a regularization-free approach
Figure 4 for Sub-8-bit quantization for on-device speech recognition: a regularization-free approach
Viaarxiv icon

Sub-8-Bit Quantization Aware Training for 8-Bit Neural Network Accelerator with On-Device Speech Recognition

Add code
Jun 30, 2022
Figure 1 for Sub-8-Bit Quantization Aware Training for 8-Bit Neural Network Accelerator with On-Device Speech Recognition
Figure 2 for Sub-8-Bit Quantization Aware Training for 8-Bit Neural Network Accelerator with On-Device Speech Recognition
Figure 3 for Sub-8-Bit Quantization Aware Training for 8-Bit Neural Network Accelerator with On-Device Speech Recognition
Figure 4 for Sub-8-Bit Quantization Aware Training for 8-Bit Neural Network Accelerator with On-Device Speech Recognition
Viaarxiv icon

Scalable and Efficient Neural Speech Coding

Add code
Mar 27, 2021
Figure 1 for Scalable and Efficient Neural Speech Coding
Figure 2 for Scalable and Efficient Neural Speech Coding
Figure 3 for Scalable and Efficient Neural Speech Coding
Figure 4 for Scalable and Efficient Neural Speech Coding
Viaarxiv icon

Sparsification via Compressed Sensing for Automatic Speech Recognition

Add code
Feb 09, 2021
Figure 1 for Sparsification via Compressed Sensing for Automatic Speech Recognition
Figure 2 for Sparsification via Compressed Sensing for Automatic Speech Recognition
Figure 3 for Sparsification via Compressed Sensing for Automatic Speech Recognition
Figure 4 for Sparsification via Compressed Sensing for Automatic Speech Recognition
Viaarxiv icon

Psychoacoustic Calibration of Loss Functions for Efficient End-to-End Neural Audio Coding

Add code
Dec 31, 2020
Figure 1 for Psychoacoustic Calibration of Loss Functions for Efficient End-to-End Neural Audio Coding
Figure 2 for Psychoacoustic Calibration of Loss Functions for Efficient End-to-End Neural Audio Coding
Figure 3 for Psychoacoustic Calibration of Loss Functions for Efficient End-to-End Neural Audio Coding
Figure 4 for Psychoacoustic Calibration of Loss Functions for Efficient End-to-End Neural Audio Coding
Viaarxiv icon

Efficient Context Aggregation for End-to-End Speech Enhancement Using a Densely Connected Convolutional and Recurrent Network

Add code
Aug 18, 2019
Figure 1 for Efficient Context Aggregation for End-to-End Speech Enhancement Using a Densely Connected Convolutional and Recurrent Network
Figure 2 for Efficient Context Aggregation for End-to-End Speech Enhancement Using a Densely Connected Convolutional and Recurrent Network
Figure 3 for Efficient Context Aggregation for End-to-End Speech Enhancement Using a Densely Connected Convolutional and Recurrent Network
Viaarxiv icon

Cascaded Cross-Module Residual Learning towards Lightweight End-to-End Speech Coding

Add code
Jun 18, 2019
Figure 1 for Cascaded Cross-Module Residual Learning towards Lightweight End-to-End Speech Coding
Figure 2 for Cascaded Cross-Module Residual Learning towards Lightweight End-to-End Speech Coding
Figure 3 for Cascaded Cross-Module Residual Learning towards Lightweight End-to-End Speech Coding
Figure 4 for Cascaded Cross-Module Residual Learning towards Lightweight End-to-End Speech Coding
Viaarxiv icon

A Hybrid Supervised-unsupervised Method on Image Topic Visualization with Convolutional Neural Network and LDA

Add code
Apr 09, 2017
Figure 1 for A Hybrid Supervised-unsupervised Method on Image Topic Visualization with Convolutional Neural Network and LDA
Figure 2 for A Hybrid Supervised-unsupervised Method on Image Topic Visualization with Convolutional Neural Network and LDA
Figure 3 for A Hybrid Supervised-unsupervised Method on Image Topic Visualization with Convolutional Neural Network and LDA
Figure 4 for A Hybrid Supervised-unsupervised Method on Image Topic Visualization with Convolutional Neural Network and LDA
Viaarxiv icon