Author: Lixing He, CUHK

This is a Repo contains all the research within Audio, Acoustic and Speech. Each Folder refers to one category of project (may not finished), splitted version will be provided as well.

Authentication

1. Bone-conduction based authentication
2. Acoustic authentication
3. Audio speaker identification

Computing

1. deployment of audio tasks on Raspberry Pi 4B (64 Bit)
2. DeepEar (localization) + VGGVOX (identification) + More....

Enhancement

1. Bone-conduction Speech enhancement
2. Fullsubnet, Conformer, VoiceFilter
3. Mobile self-supervised framwork

Simulation

1. STL format simulation with Pyroomacoustic (ray-tracing)

Raspberry-Pi

1. I2C BMI-160 imu data loading